Skip to main content

TECNOLOGIAS DE APRENDIZAJE PROFUNDO APLICADAS AL PROCESADO DE VOZ Y AUDIO

TEC2015-69266-P

•
Funding agency name Ministerio de Economía y Competitividad
Funding agency acronym MINECO
Program Programa Estatal de Fomento de la Investigación Científica y Técnica de Excelencia
Subprogram Subprograma Estatal de Generación de Conocimiento
Call Proyectos de I+D (Excelencia / Generación de Conocimiento)
Call year 2015
State Plan 2013-2016 Plan Estatal de Investigación Científica y Técnica y de Innovación
Management unit Secretaría de Estado de Educación, Formación Profesional y Universidades
Beneficiary centre UNIVERSITAT POLITECNICA DE CATALUNYA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 101
Encontrada(s) 2 página(s)

Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Towards large scale multimedia indexing: a case study on person discovery in broadcast news

  • Le, Nam
  • Bredin, Hervé
  • Sargent, Gabriel
  • India, Miquel
  • López Otero, Paula
  • Barras, Claude
  • Guinaudeau, Camille
  • Gravier, Guillaume
  • Barbosa da Fonseca, Gabriel
  • Lyon Freire, Izabela
  • Patrocínio, Zenilton
  • Guimarães, Silvio Jamil F.
  • Martí Juan, Gerard
  • Morros, Josep Ramon
  • Hernando, Javier
  • Docio Fernández, Laura
  • García Mateo, Carmen
  • Meignier, Sylvain
  • Odobez, Jean-Marc
Comunicació presentada a: the 15th International Workshop on Content-Based Multimedia Indexing (CBMI'17), celebrat a Florència, Itàlia, del 19 al 21 de juny de 2017, The rapid growth of multimedia databases and the human interest in their peers make indices representing the location and identity of people in audio-visual documents essential for searching archives. Person discovery in the absence of prior identity knowledge requires accurate association of audio-visual cues and detected names. To this end, we present 3 different strategies to approach this problem: clustering-based naming, verification-based naming, and graph-based naming. Each of these strategies utilizes different recent advances in unsupervised face / speech representation, verification, and optimization. To have a better understanding of the approaches, this paper also provides a quantitative and qualitative comparative study of these approaches using the associated corpus of the Person Discovery challenge at MediaEval 2016. From the results of our experiments, we can observe the pros and cons of each approach, thus paving the way for future promising research directions., This work was supported by the EU project EUMSSI (FP7-611057), ANR project MetaDaTV (ANR-14-CE24-0024) project, Camomile project (PCIN-2013-067), and the projects TEC2013-43935-R, TEC2015-69266-P, TEC2016-75976-R, TEC2015-65345-P financed by the Spanish government and ERDF.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

CATOTRON – A Neural text-to-speech system in catalan

  • Külebi, Baybars
  • Öktem, Alp
  • Peiró Lilja, Àlex
  • Pascual, Santiago
  • Farrús, Mireia
Comunicació presentada a Interspeech 2020 celebrat del 25 al 29 d'octubre de 2020 a Shanghai, Xina., We present Catotron, a neural network-based open-source speech synthesis system in Catalan. Catotron consists of a sequence-to-sequence model trained with two small opensource datasets based on semi-spontaneous and read speech. We demonstrate how a neural TTS can be built for languages with limited resources using found-data optimization and crosslingual transfer learning. We make the datasets, initial models and source code publicly available for both commercial and research purposes., This work was subsidised by the Catalan Department of Culture. A part of the funding comes from the financing administered by the inheritance board of the Generalitat de Catalunya. The last author has been funded by the Agencia Estatal de Investigaci ´on (AEI), Ministerio de Ciencia, Innovaci´on y Universidades and the Fondo Social Europeo (FSE) under grant RYC- 2015-17239 (AEI/FSE, UE). Part of this work has been carried out using an NVIDIA GPU Titan Xp generously provided by NVIDIA Company. This research was also partially supported by the project TEC2015-69266-P (MINECO/FEDER, UE).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

A Deep Source-Context Feature for Lexical Selection in Statistical Machine Translation

  • Gupta, Parth Alokkumar
  • Costa-Jussa, Marta R
  • Banchs, Rafael
  • Rosso, Paolo|||0000-0002-8922-1242
This paper presents a methodology to address lexical disambiguation in a standard phrase-based statistical machine translation system. Similarity among source contexts is used to select appropriate translation units. The information is introduced as a novel feature of the phrase-based model and it is used to select the translation units extracted from the training sentence more similar to the sentence to translate. The similarity is computed through a deep autoencoder representation, which allows to obtain effective lowdimensional embedding of data and statistically significant BLEU score improvements on two different tasks (English-to-Spanish and English-to-Hindi). © 2016 Elsevier B.V. All rights reserved., The work of the first author has been supported by FPI UPV pre-doctoral grant (num. registro - 3505). The work of the second author has been supported by Spanish Ministerio de Economia y Competitividad, contract TEC2015-69266-P and the Seventh Framework Program of the European Commission through the International Outgoing Fellowship Marie Curie Action (IMTraP-2011-29951). The work of the third author has been supported by the Spanish Ministerio de Economia y Competitividad, SomEMBED TIN2015-71147-C2-1-P research project and by the Generalitat Valenciana under the grant ALMAPATER (PrometeoII/2014/030).
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Generación morfológica con algoritmos de aprendizaje profundo integrada en un sistema de traducción automática estadística, Integration of morphology generation techniques based on deep learning into a statistical machine translation system

  • Escolano, Carlos
  • Ruiz Costa-Jussà, Marta
La variación morfológica entre un lenguaje fuente y el lenguaje destino genera dificultades a los algoritmos estándares de traducción como el estadístico basado en segmentos. En este trabajo planteamos dividir la tarea de traducción en dos partes: primero, simplificamos el lenguaje destino en términos morfológicos y construimos el sistema de traducción con esta modificación; y después utilizamos un algoritmo de clasificación para generar la morfología final. Este trabajo presenta una arquitectura de aprendizaje profundo que permite añadir de manera efectiva la información morfológica a la traducción simplificada generada por un traductor estadístico basado en segmentos. Demostramos que la arquitectura diseñada presenta resultados superiores a los algoritmos estado-del-arte en términos de precisión y que la calidad de la traducción mejora en términos de METEOR.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Coverage for Character Based Neural Machine Translation, Técnicas de Cobertura aplicadas al Sistema de Traducción Automática Neuronal basado en Caracteres

  • Kazimi, M. Bashir
  • Ruiz Costa-Jussà, Marta
In recent years, Neural Machine Translation (NMT) has achieved state-of-the-art performance in translating from a language; source language, to another; target language. However, many of the proposed methods use word embedding techniques to represent a sentence in the source or target language. Character embedding techniques for this task has been suggested to represent the words in a sentence better. Moreover, recent NMT models use attention mechanism where the most relevant words in a source sentence are used to generate a target word. The problem with this approach is that while some words are translated multiple times, some other words are not translated. To address this problem, coverage model has been integrated into NMT to keep track of already-translated words and focus on the untranslated ones. In this research, we present a new architecture in which we use character embedding for representing the source and target languages, and also use coverage model to make certain that all words are translated. Experiments were performed to compare our model with coverage and character model and the results show that our model performs better than the other two models.
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Auto-encoding nearest neighbor i-vectors for speaker verification

  • Khan, Umair
  • India Massana, Miquel Àngel|||0000-0002-3107-3662
  • Hernando Pericás, Francisco Javier|||0000-0002-1730-8154
In the last years, i-vectors followed by cosine or PLDA scoringtechniques were the state-of-the-art approach in speaker veri-fication. PLDA requires labeled background data, and thereexists a significant performance gap between the two scoringtechniques. In this work, we propose to reduce this gap by us-ing an autoencoder to transform i-vector into a new speaker vec-tor representation, which will be referred to as ae-vector. Theautoencoder will be trained to reconstruct neighbor i-vectors in-stead of the same training i-vectors, as usual. These neighbori-vectors will be selected in an unsupervised manner accordingto the highest cosine scores to the training i-vectors. The evalua-tion is performed on the speaker verification trials of VoxCeleb-1 database. The experiments show that our proposed ae-vectorsgain a relative improvement of 42% in terms of EER comparedto the conventional i-vectors using cosine scoring, which fillsthe performance gap between cosine and PLDA scoring tech-niques by 92%, but without using speaker labels, Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Exploring efficient neural architectures for linguistic-acoustic mapping in text-to-speech

  • Pascual de la Puente, Santiago|||0000-0002-8365-7387
  • Serra, Joan
  • Bonafonte Cávez, Antonio|||0000-0002-6240-9915
Conversion from text to speech relies on the accurate mapping from linguistic to acoustic symbol sequences, for which current practice employs recurrent statistical models such as recurrent neural networks. Despite the good performance of such models (in terms of low distortion in the generated speech), their recursive structure with intermediate affine transformations tends to make them slow to train and to sample from. In this work, we explore two different mechanisms that enhance the operational efficiency of recurrent neural networks, and study their performance–speed trade-off. The first mechanism is based on the quasi-recurrent neural network, where expensive affine transformations are removed from temporal connections and placed only on feed-forward computational directions. The second mechanism includes a module based on the transformer decoder network, designed without recurrent connections but emulating them with attention and positioning codes. Our results show that the proposed decoder networks are competitive in terms of distortion when compared to a recurrent baseline, whilst being significantly faster in terms of CPU and GPU inference time. The best performing model is the one based on the quasi-recurrent mechanism, reaching the same level of naturalness as the recurrent neural network based model with a speedup of 11.2 on CPU and 3.3 on GPU., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Estudio, informe, memoria (Report)

Automatic detection of alarm sounds in a noisy hospital environment using model and non-model based approaches

  • Raboshchuk, Ganna
  • Gómez Quintana, Sergi
  • Peiró Lilja, Alexandre
  • Nadeu Camprubí, Climent|||0000-0002-5863-0983
Article publicat sense revisió per parells a Arxiv, In the noisy acoustic environment of a Neonatal Intensive Care Unit (NICU) there is a variety of alarms, which are frequently triggered by the biomedical equipment. In this paper different approaches for automatic detection of those sound alarms are presented and compared: 1) a non-model-based approach that employs signal processing techniques; 2) a model-based approach based on neural networks; and 3) an approach that combines both non-model and model-based approaches. The performance of the developed detection systems that follow each of those approaches is assessed, analysed and compared both at the frame level and at the event level by using an audio database recorded in a real-world hospital environment.
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

A geometric proxy of economic uncertainty based on the disagreement in survey expectations

  • Claveria González, Oscar
  • Monte Moreno, Enrique|||0000-0002-4907-0494
  • Torra Porras, Salvador
In this study we present a geometric approach to proxy economic uncertainty. We design a positional indicator of disagreement among survey-based agents' expectations about the state of the economy. Previous dispersion-based uncertainty measures derived from business and consumer surveys exclusively make use of the two extreme pieces of information: the percentage of respondents expecting a variable to rise and to fall. With the aim of also incorporating the information coming from the share of respondents expecting a variable to remain constant, we propose a geometrical framework and use a barycentric coordinate system to generate a metric of disagreement, referred to as a discrepancy indicator. We assess its performance, both empirically and experimentally, by comparing it to the standard deviation of the share of positive and negative responses, which has been used by Bachman et al. (2013) as a proxy for economic uncertainty. When applied in sixteen European countries, we find that both time-varying metrics co-evolve in most countries for expectations about the country's overall economic situation in the present, but not in the future. Additionally, we obtain their simulated sampling distributions and we find that the proposed indicator gravitates uniformly towards the three vertices of the simplex representing the three answering categories, as opposed to the standard deviation, which tends to overestimate the level of uncertainty as a result of ignoring the no-change responses. Consequently, we find evidence that the information coming from agents expecting a variable to remain constant has an effect on the measurement of disagreement., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Spanish statistical parametric speech synthesis using a neural vocoder

  • Bonafonte Cávez, Antonio|||0000-0002-6240-9915
  • Pascual de la Puente, Santiago|||0000-0002-8365-7387
  • Dorca, G.
During the 2000s decade, unit-selection based text-to-speech was the dominant commercial technology. Meanwhile, the TTS research community has made a big effort to push statistical-parametric speech synthesis to get similar quality and more flexibility on the synthetically generated voice. During last years, deep learning advances applied to speech synthesis have filled the gap, specially when neural vocoders substitute traditional signal-processing based vocoders. In this paper we propose to substitute the waveform generation vocoder of MUSA, our Spanish TTS, with SampleRNN, a neural vocoder which was recently proposed as a deep autoregressive raw waveform generation model. MUSA uses recurrent neural networks to predict vocoder parameters (MFCC and logF0) from linguistic features. Then, the Ahocoder vocoder is used to recover the speech waveform out of the predicted parameters. In the first system SampleRNN is extended to generate speech conditioned on the Ahocoder generated parameters (mfcc and logF0), where two configurations have been considered to train the system. First, the parameters derived from the signal using Ahocoder are used. Secondly, the system is trained with the parameters predicted by MUSA, where SampleRNN and MUSA are jointly optimized. The subjective evaluation shows that the second system outperforms both the original Ahocoder and SampleRNN as an independent neural vocoder., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Language modelling for speaker diarization in telephonic interviews

  • India Massana, Miquel Àngel|||0000-0002-3107-3662
  • Hernando Pericás, Francisco Javier|||0000-0002-1730-8154
  • Rodríguez Fonollosa, José Adrián|||0000-0001-9513-7939
The aim of this paper is to investigate the benefit of combining both language and acoustic modelling for speaker diarization. Although conventional systems only use acoustic features, in some scenarios linguistic data contain high discriminative speaker information, even more reliable than the acoustic ones. In this study we analyze how an appropriate fusion of both kind of features is able to obtain good results in these cases. The proposed system is based on an iterative algorithm where a LSTM network is used as a speaker classifier. The network is fed with character-level word embeddings and a GMM based acoustic score created with the output labels from previous iterations. The presented algorithm has been evaluated in a Call-Center database, which is composed of telephone interview audios. The combination of acoustic features and linguistic content shows a 84.29% improvement in terms of a word-level DER as compared to a HMM/VB baseline system. The results of this study confirms that linguistic content can be efficiently used for some speaker recognition tasks., This work was partially supported by the Spanish Project DeepVoice (TEC2015-69266-P) and by the project PID2019-107579RBI00/ AEI /10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Evaluating the underlying gender bias in contextualized word embeddings

  • Basta, Christine Raouf Saad
  • Ruiz Costa-Jussà, Marta|||0000-0002-5703-520X
  • Casas Manzanares, Noé
Gender bias is highly impacting natural language processing applications. Word embeddings have clearly been proven both to keep and amplify gender biases that are present in current data sources. Recently, contextualized word embeddings have enhanced previous word embedding techniques by computing word vector representations dependent on the sentence they appear in. In this paper, we study the impact of this conceptual change in the word embedding computation in relation with gender bias. Our analysis includes different measures previously applied in the literature to standard word embeddings. Our findings suggest that contextualized word embeddings are less biased than standard ones even when the latter are debiased., We want to thank Hila Gonen for her support dur-ing our research.This work is supported in part by the Catalan Agency for Management of University andResearch Grants (AGAUR) through the FI PhDScholarship and the Industrial PhD Grant. Thiswork is also supported in part by the Span-ish Ministerio de Economa y Competitividad, the European Regional Development Fund andthe Agencia Estatal de Investigacin, through thepostdoctoral senior grant Ramn y Cajal, contract TEC2015-69266-P (MINECO/FEDER,EU)and contract PCIN-2017-079 (AEI/MINECO)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Self multi-head attention for speaker recognition

  • India Massana, Miquel Àngel|||0000-0002-3107-3662
  • Safari, Pooyan
  • Hernando Pericás, Francisco Javier|||0000-0002-1730-8154
Most state-of-the-art Deep Learning (DL) approaches forspeaker recognition work on a short utterance level. Given thespeech signal, these algorithms extract a sequence of speakerembeddings from short segments and those are averaged to ob-tain an utterance level speaker representation. In this work wepropose the use of an attention mechanism to obtain a discrim-inative speaker embedding given non fixed length speech utter-ances. Our system is based on a Convolutional Neural Network(CNN) that encodes short-term speaker features from the spec-trogram and a self multi-head attention model that maps theserepresentations into a long-term speaker embedding. The atten-tion model that we propose produces multiple alignments fromdifferent subsegments of the CNN encoded states over the se-quence. Hence this mechanism works as a pooling layer whichdecides the most discriminative features over the sequence toobtain an utterance level representation. We have tested thisapproach for the verification task for the VoxCeleb1 dataset.The results show that self multi-head attention outperforms bothtemporal and statistical pooling methods with a18%of rela-tive EER. Obtained results show a58%relative improvementin EER compared to i-vector+PLDA, Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

A novel approach to real-time range estimation of underwater acoustic sources using supervised machine learning

  • Houégnigan, Ludwig|||0000-0003-2903-3140
  • Safari, Pooyan
  • Nadeu Camprubí, Climent|||0000-0002-5863-0983
  • Van der Schaar, Mike Connor Roger Malcolm
  • André, Michel|||0000-0002-0091-7279
© 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., The proposed paper introduces a novel method for range estimation of acoustic sources, both cetaceans and industrial sources, in deep sea environments using supervised learning with neural networks in the contex of a single sensor, a compact array, or a small aperture towed array. The presented results have potential both for industrial impact and for the conservation and density estimation of cetaceans. With an average error of 4.3% for ranges up to 8 kilometers and typically below 300 meters, those results are challenging and to our knowledge they are unprecedented for an automated real-time solution., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Byte-based neural machine translation

  • Ruiz Costa-Jussà, Marta|||0000-0002-5703-520X
  • Escolano Peinado, Carlos|||0000-0001-6657-673X
  • Rodríguez Fonollosa, José Adrián|||0000-0001-9513-7939
This paper presents experiments compar- ing character-based and byte-based neural machine translation systems. The main motivation of the byte-based neural ma- chine translation system is to build multi- lingual neural machine translation systems that can share the same vocabulary. We compare the performance of both systems in several language pairs and we see that the performance in test is similar for most language pairs while the training time is slightly reduced in the case of byte-based neural machine translation.
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

TEC forecasting based on manifold trajectories

  • Monte Moreno, Enrique|||0000-0002-4907-0494
  • García Rigo, Alberto|||0000-0001-6226-4851
  • Hernández Pajares, Manuel|||0000-0002-9687-5850
  • Yang, Heng
In this paper, we present a method for forecasting the ionospheric Total Electron Content (TEC) distribution from the International GNSS Service’s Global Ionospheric Maps. The forecasting system gives an estimation of the value of the TEC distribution based on linear combination of previous TEC maps (i.e., a set of 2D arrays indexed by time), and the computation of a tangent subspace in a manifold associated to each map. The use of the tangent space to each map is justified because it allows modeling the possible distortions from one observation to the next as a trajectory on the tangent manifold of the map. The coefficients of the linear combination of the last observations along with the tangent space are estimated at each time stamp to minimize the mean square forecasting error with a regularization term. The estimation is made at each time stamp to adapt the forecast to short-term variations in solar activity., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Terminology-aware segmentation and domain feature for the WMT19 biomedical translation task

  • Carrino, Casimiro Pio
  • Rafieian, Bardia
  • Ruiz Costa-Jussà, Marta|||0000-0002-5703-520X
  • Rodríguez Fonollosa, José Adrián|||0000-0001-9513-7939
In this work, we give a description of the TALP-UPC systems submitted for the WMT19 Biomedical Translation Task. Our proposed strategy is NMT model-independent and relies only on one ingredient, a biomedical terminology list. We first extracted such a terminology list by labelling biomedical words in our training dataset using the BabelNet API. Then, we designed a data preparation strategy to insert the terms information at a token level. Finally, we trained the Transformer model with this terms-informed data. Our best-submitted system ranked 2nd and 3rd for Spanish-English and English-Spanish translation directions, respectively., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Estudio, informe, memoria (Report)

A geometric approach to proxy economic uncertainty by a metric of disagreement among qualitative expectations

  • Claveria González, Oscar
  • Monte Moreno, Enrique|||0000-0002-4907-0494
  • Torra Porras, Salvador
Working paper, In this study we present a geometric approach to proxy economic uncertainty. We design a positional indicator of disagreement among survey-based agents' expectations about the state of the economy. Previous dispersion-based uncertainty indicators derived from business and consumer surveys exclusively make use of the two extreme pieces of information coming from the respondents expecting a variable to rise and to fall. With the aim of also incorporating the information coming from the share of respondents expecting a variable to remain constant, we propose a geometrical framework and use a barycentric coordinate system to generate a measure of disagreement, referred to as a discrepancy indicator. We assess its performance, both empirically and experimentally, by comparing it to the standard deviation of the share of positive and negative responses, which has been used by Bachman et al. (2013) as a proxy for economic uncertainty. When applied in sixteen European countries, we find that both time-varying metrics co-evolve in most countries for expectations about the country's overall economic situation in the present, but not in the future. Additionally, we obtain their simulated sampling distributions and we find that the proposed indicator gravitates uniformly towards the three vertices of the simplex representing the three answering categories, as opposed to the standard deviation, which tends to overestimate the level of uncertainty as a result of ignoring the no-change responses. Consequently, we find evidence that the information coming from agents expecting a variable to remain constant has an effect on the measurement of disagreement.
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Detection and description of the different ionospheric disturbances that appeared during the solar eclipse of 21 August 2017

  • Yang, Heng
  • Monte Moreno, Enrique|||0000-0002-4907-0494
  • Hernández Pajares, Manuel|||0000-0002-9687-5850
This work will provide a detailed characterization of the travelling ionospheric disturbances (TIDs) created by the solar eclipse of 21 August 2017, the shadow of which crossed the United States from the Pacific to the Atlantic ocean. The analysis is done by means of the Atomic Decomposition Detector of Traveling Ionospheric Disturbances (ADDTID) algorithm. This method automatically detects and characterizes multiple TIDs from the global navigation satellite system (GNSS) observation. The set of disturbances generated by the eclipse has a richer and more varied behavior than that associated with the shock wave directly produced by cooling effects of the moon shadow. This can be modeled in part as if the umbra and penumbra of the eclipse were moving cylinders that intersects with variable elevation angle a curved surface. This projection gives rise to regions of equal penumbra with shapes similar to ellipses, with different centers and foci. The result of this is reflected in the time evolution of the TID wavelengths produced by the eclipse, which depend on the vertical angle of the sun with the surface of the earth, and also a double bow wave phenomenon, where the bow waves are generated in advance to the umbra. We show that the delay in the appearance of the disturbances with the transit of the eclipse are compatible with the physical explanations, linked to the different origins of the disturbances and the wavelengths. Finally, we detected a consistent pattern, in location and time of disturbances in advance to the penumbra as a set of medium scale TIDs, which could be hypothesized as soliton waves of the bow wave. In all cases, the detected disturbances were checked visually on the detrended vertical total electron content (TEC) maps., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Empirical modelling of survey-based expectations for the design of economic indicators in five European regions

  • Claveria, Oscar
  • Monte Moreno, Enrique|||0000-0002-4907-0494
  • Torra Porras, Salvador
This is a post-peer-review, pre-copyedit version of an article published in Empirica. The final authenticated version is available online at: http://dx.doi.org/10.1007/s10663-017-9395-1”., In this study we use agents’ expectations about the state of the economy to generate indicators of economic activity in twenty-six European countries grouped in five regions (Western, Eastern, and Southern Europe, and Baltic and Scandinavian countries). We apply a data-driven procedure based on evolutionary computation to transform survey variables in economic growth rates. In a first step, we design five independent experiments to derive a formula using survey variables that best replicates the evolution of economic growth in each region by means of genetic programming, limiting the integration schemes to the main mathematical operations. We then rank survey variables according to their performance in tracking economic activity, finding that agents’ ‘‘perception about the overall economy compared to last year’’ is the survey variable with the highest predictive power. In a second step, we assess the out-of-sample forecast accuracy of the evolved indicators. Although we obtain different results across regions, Austria, Slovakia, Portugal, Lithuania and Sweden are the economies of each region that show the best forecast results. We also find evidence that the forecasting performance of the survey-based indicators improves during periods of higher growth., This is a post-peer-review, pre-copyedit version of an article published in Empirica. The final authenticated version is available online at: http://dx.doi.org/10.1007/s10663-017-9395-1”., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Multilingual and interlingual semantic representations for natural language processing: a brief introduction

  • Ruiz Costa-Jussà, Marta|||0000-0002-5703-520X
  • España-Bonet, Cristina
  • Fung, Pascale
  • Smith, Noah
We introduce the Computational Linguistics special issue on Multilingual and Interlingual Semantic Representations for Natural Language Processing. We situate the special issue’s five articles in the context of our fast-changing field, explaining our motivation for this project. We offer a brief summary of the work in the issue, which includes developments on lexical and sentential semantic representations, from symbolic and neural perspectives., MRC is supported in part by a Google Faculty Research Award 2018, Spanish Ministerio de Economía y Competitividad, the European Regional Development Fund and the Agencia Estatal de Investigacion, through the postdoctoral senior grant Ramon y Cajal, the contract TEC2015-69266-P (MINECO/FEDER,EU) and the contract PCIN-2017-079 (AEI/MINECO). N. A. S. is supported by National Science Foundation grant IIS-1562364. C. E. B. is funded by the German Federal Ministry of Education and Research under the funding code 01IW17001 (Deeplee)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

DNN speaker embeddings using autoencoder pre-training

  • Khan, Umair
  • Hernando Pericás, Francisco Javier|||0000-0002-1730-8154
Over the last years, i-vectors have been the state-of-the-art approach in speaker recognition. Recent improvements in deep learning have increased the discriminative quality of i-vectors. However, deep learning architectures require a large amount of labeled background data which is difficult in practice. The aim of this paper is to propose an alternative scheme in order to reduce the need of labeled data. We propose the use of autoencoder pre-training in a speaker verification task. First, we train an autoencoder in an unsupervised way, using a large amount of unlabeled background data. Then, we train a Deep Neural Network (DNN) initialized with the parameters of the pre-trained autoencoder. The DNN training is carried out in a supervised way using relatively small labeled background data. In the testing phase, we extract speaker embeddings as the output of an intermediate layer of the DNN. The training and evaluation were performed on VoxCeleb-2 and VoxCeleb1 databases, respectively. The experimental results have shown that by initializing DNN with the parameters of the pre-trained autoencoder, we have achieved a relative improvement of 21%, in terms of Equal Error Rate (EER), over the baseline i-vector/PLDA system., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

I-vector transformation using k-nearest neighbors for speaker verification

  • Khan, Umair
  • India Massana, Miquel Àngel|||0000-0002-3107-3662
  • Hernando Pericás, Francisco Javier|||0000-0002-1730-8154
© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. https://dx.doi.org/10.1109/ICASSP40776.2020.9053504, Probabilistic Linear Discriminant Analysis (PLDA) is the most efficient backend for i-vectors. However, it requires labeled background data which can be difficult to access in practice. Unlike PLDA, cosine scoring avoids speaker-labels at the cost of degrading the performance. In this work, we propose a post processing of i-vectors using a Deep Neural Network (DNN) to transform i-vectors into a new speaker vector representation. The DNN will be trained using i-vectors that are similar to the training i-vectors. These similar i-vectors will be selected in an unsupervised manner. Using the new vector representation, we will score the experimental trials using cosine scoring. The evaluation was performed on the speaker verification trials of VoxCeleb-1 database. The experiments have shown that with the help of the similar i-vectors the new vectors become more discriminative than the original i-vectors. The new vectors have gained a relative improvement of 53% in terms of EER, compared to the conventional i-vector/PLDA system, but without using speaker labels., This work has been developed in the framework of DeepVoice Project (TEC2015-69266-P), funded by Spanish Ministry., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Prosodic and spectral iVectors for expressive speech synthesis

  • Jauk, Igor
  • Bonafonte Cávez, Antonio|||0000-0002-6240-9915
This work presents a study on the suitability of prosodic andacoustic features, with a special focus on i-vectors, in expressivespeech analysis and synthesis. For each utterance of two dif-ferent databases, a laboratory recorded emotional acted speech,and an audiobook, several prosodic and acoustic features are ex-tracted. Among them, i-vectors are built not only on the MFCCbase, but also on F0, power and syllable durations. Then, un-supervised clustering is performed using different feature com-binations. The resulting clusters are evaluated calculating clus-ter entropy for labeled portions of the databases. Additionally,synthetic voices are trained, applying speaker adaptive training,from the clusters built from the audiobook. The voices are eval-uated in a perceptual test where the participants have to edit anaudiobook paragraph using the synthetic voices.The objective results suggest that i-vectors are very use-ful for the audiobook, where different speakers (book charac-ters) are imitated. On the other hand, for the laboratory record-ings, traditional prosodic features outperform i-vectors. Also,a closer analysis of the created clusters suggest that differentspeakers use different prosodic and acoustic means to conveyemotions. The perceptual results suggest that the proposed i-vector based feature combinations can be used for audiobookclustering and voice training., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Automatic speech recognition with deep neural networks for impaired speech

  • España-i-Bonet, Cristina
  • Rodríguez Fonollosa, José Adrián|||0000-0001-9513-7939
The final publication is available at https://link.springer.com/chapter/10.1007%2F978-3-319-49169-1_10, Automatic Speech Recognition has reached almost human performance in some controlled scenarios. However, recognition of impaired speech is a difficult task for two main reasons: data is (i) scarce and (ii) heterogeneous. In this work we train different architectures on a database of dysarthric speech. A comparison between architectures shows that, even with a small database, hybrid DNN-HMM models outperform classical GMM-HMM according to word error rate measures. A DNN is able to improve the recognition word error rate a 13% for subjects with dysarthria with respect to the best classical architecture. This improvement is higher than the one given by other deep neural networks such as CNNs, TDNNs and LSTMs. All the experiments have been done with the Kaldi toolkit for speech recognition for which we have adapted several recipes to deal with dysarthric speech and work on the TORGO database. These recipes are publicly available., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Multilingual, multi-scale and multi-layer visualization of sequence-based intermediate representations

  • Escolano Peinado, Carlos|||0000-0001-6657-673X
  • Ruiz Costa-Jussà, Marta|||0000-0002-5703-520X
  • Lacroux, Elora
  • Vázquez Alcocer, Pere Pau|||0000-0003-4638-4065
The main alternatives nowadays to dealwith sequences are Recurrent Neural Net-works (RNN), Convolutional Neural Networks(CNN) architectures and the Transformer. Inthis context, RNN’s, CNN’s and Transformerhave most commonly been used as an encoder-decoder architecture with multiple layers ineach module. Far beyond this, these architec-tures are the basis for the contextual word em-beddings which are revolutionizing most natural language downstream applications. However, intermediate layer representations insequence-based architectures can be difficultto interpret. To make each layer representation within these architectures more accessible and meaningful, we introduce a web-based toolthat visualizes them both at the sentence and token level. We present three use cases. The first analyses gender issues in contextual worde mbeddings. The second and third are show-ing multilingual intermediate representations for sentences and tokens and the evolution of these intermediate representations along the multiple layers of the decoder and in the con-text of multilingual machine translation., This work is supported by a Google Faculty Research Award. This workis also supported by the Spanish Ministerio de Economía y Competitividad, the European Regional Development Fund and the Agencia Estatal de Investigación, through the post-doctoral senior grant Ramón y Cajal, contracts TEC2015-69266-P and TIN2017-88515-C2-1-R(GEN3DLIVE) (MINECO/FEDER,EU), and contract PCIN-2017-079 (AEI/MINECO)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

From bilingual to multilingual neural machine translation by incremental training

  • Escolano Peinado, Carlos|||0000-0001-6657-673X
  • Ruiz Costa-Jussà, Marta|||0000-0002-5703-520X
  • Rodríguez Fonollosa, José Adrián|||0000-0001-9513-7939
Multilingual Neural Machine Translation approaches are based on the use of task specific models and the addition of one more language can only be done by retraining the whole system. In this work, we propose a new training schedule that allows the system to scale to more languages without modification of the previous components based on joint training and language-independent encoder/decoder modules allowing for zero-shot translation. This work in progress shows close results to state-of-the-art in the WMT task., This work is supported in part by a Google Faculty Research Award. This work is also supported in part by the Spanish Ministerio de Economa y Competitividad, the European Regional Development Fund and the Agencia Estatal de Investigacin, through the postdoctoral senior grant Ramon y Cajal, contract TEC2015-69266-P (MINECO/FEDER,EU) and contract PCIN-2017- 079 (AEI/MINECO)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Self-attention encoding and pooling for speaker recognition

  • Safari, Pooyan
  • India Massana, Miquel Àngel|||0000-0002-3107-3662
  • Hernando Pericás, Francisco Javier|||0000-0002-1730-8154
The computing power of mobile devices limits the end-user applications in terms of storage size, processing, memory and energy consumption. These limitations motivate researchers for the design of more efficient deep models. On the other hand, self-attention networks based on Transformer architecture have attracted remarkable interests due to their high parallelization capabilities and strong performance on a variety of Natural Language Processing (NLP) applications. Inspired by the Transformer, we propose a tandem Self-Attention Encoding and Pooling (SAEP) mechanism to obtain a discriminative speaker embedding given non-fixed length speech utterances. SAEP is a stack of identical blocks solely relied on self-attention and position-wise feed-forward networks to create vector representation of speakers. This approach encodes short-term speaker spectral features into speaker embeddings to be used in text-independent speaker verification. We have evaluated this approach on both VoxCeleb1 & 2 datasets. The proposed architecture is able to outperform the baseline x-vector, and shows competitive performance to some other benchmarks based on convolutions, with a significant reduction in model size. It employs 94%, 95%, and 73% less parameters compared to ResNet-34, ResNet-50, and x-vector, respectively. This indicates that the proposed fully attention based architecture is more efficient in extracting time-invariant features from speaker utterances., This work was supported in part by the Spanish Project DeepVoice (TEC2015-69266-P)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Evolutionary computation for macroeconomic forecasting

  • Claveria, Oscar
  • Monte Moreno, Enrique|||0000-0002-4907-0494
  • Torra Porras, Salvador
The final publication is available at Springer via http://dx.doi.org/10.1007/s10614-017-9767-4, The main objective of this study is twofold. First, we propose an empirical modelling approach based on genetic programming to forecast economic growth by means of survey data on expectations. We use evolutionary algorithms to estimate a symbolic regression that links survey-based expectations to a quantitative variable used as a yardstick, deriving mathematical functional forms that approximate the target variable. The set of empirically-generated proxies of economic growth are used as building blocks to forecast the evolution of GDP. Second, we use these estimates of GDP to assess the impact of the 2008 financial crisis on the accuracy of agents’ expectations about the evolution of the economic activity in four Scandinavian economies. While we find an improvement in the capacity of agents’ to anticipate economic growth after the crisis, predictive accuracy worsens in relation to the period prior to the crisis. The most accurate GDP forecasts are obtained for Sweden., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Problem-agnostic speech embeddings for multi-speaker text-to-speech with SampleRNN

  • Álvarez Robert, David
  • Pascual de la Puente, Santiago|||0000-0002-8365-7387
  • Bonafonte Cávez, Antonio|||0000-0002-6240-9915
Text-to-speech (TTS) acoustic models map linguistic features into an acoustic representation out of which an audible waveform is generated. The latest and most natural TTS systems build a direct mapping between linguistic and waveform domains, like SampleRNN. This way, possible signal naturalness losses are avoided as intermediate acoustic representations are discarded. Another important dimension of study apart from naturalness is their adaptability to generate voice from new speakers that were unseen during training. In this paper we first propose the use of problem-agnostic speech embeddings in a multi-speaker acoustic model for TTS based on SampleRNN. This way, we feed the acoustic model with speaker acousticallydependent representations that enrich the waveform generation more than embeddings unrelated to these factors. Our first results suggest that the proposed embeddings lead to better quality voices than those obtained with one-hot embeddings. Furthermore, as we can use any speech segment as an encoded representation during inference, the model is capable to generalize to new speaker identities without retraining the network. We finally show that, with a small increase of speech duration in the embedding extractor, we dramatically reduce the spectral distortion to close the gap towards the target identities., This research was supported by the project TEC2015-69266-P (MINECO/FEDER, UE)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Estudio, informe, memoria (Report)

A regional perspective on the accuracy of machine learning forecasts of tourism demand based on data characteristics

  • Claveria González, Oscar
  • Monte Moreno, Enrique|||0000-0002-4907-0494
  • Torra Porras, Salvador
Working paper, In this work we assess the role of data characteristics in the accuracy of machine learning (ML) tourism forecasts from a spatial perspective. First, we apply a seasonal-trend decomposition procedure based on non-parametric regression to isolate the different components of the time series of international tourism demand to all Spanish regions. This approach allows us to compute a set of measures to describe the features of the data. Second, we analyse the performance of several ML models in a recursive multiple-step-ahead forecasting experiment. In a third step, we rank all seventeen regions according to their characteristics and the obtained forecasting performance, and use the rankings as the input for a multivariate analysis to evaluate the interactions between time series features and the accuracy of the predictions. By means of dimensionality reduction techniques we summarise all the information into two components and project all Spanish regions into perceptual maps. We find that entropy and dispersion show a negative relation with accuracy, while the effect of other data characteristics on forecast accuracy is heavily dependent on the forecast horizon.
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Restricted Boltzmann Machine vectors for speaker clustering

  • Khan, Umair
  • Safari, Pooyan
  • Hernando Pericás, Francisco Javier|||0000-0002-1730-8154
Restricted Boltzmann Machines (RBMs) have been used both in the front-end and backend of speaker verification systems. In this work, we apply RBMs as a front-end in the context of speaker clustering. Speakers' utterances are transformed into a vector representation by means of RBMs. These vectors, referred to as RBM vectors, have shown to preserve speaker-specific information and are used for the task of speaker clustering. In this work, we perform the traditional bottom-up Agglomerative Hierarchical Clustering (AHC). Using the RBM vector representation of speakers, the performance of speaker clustering is improved. The evaluation has been performed on the audio recordings of Catalan TV Broadcast shows. The experimental results show that our proposed system outperforms the baseline i-vectors system in terms of Equal Impurity (EI). Using cosine scoring, a relative improvement of 11% and 12% are achieved for average and single linkage clustering algorithms respectively. Using PLDA scoring, the RBM vectors achieve a relative improvement of 11% compared to i-vectors for the single linkage algorithm., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

A differentiable BLEU loss. Analysis and first results

  • Casas Manzanares, Noé
  • Rodríguez Fonollosa, José Adrián|||0000-0001-9513-7939
  • Ruiz Costa-Jussà, Marta|||0000-0002-5703-520X
In natural language generation tasks, like neural machine translation and image captioning, there is usually a mismatch between the optimized loss and the de facto evaluation criterion, namely token-level maximum likelihood and corpus-level BLEU score. This article tries to reduce this gap by defining differentiable computations of the BLEU and GLEU scores. We test this approach on simple tasks, obtaining valuable lessons on its potential applications but also its pitfalls, mainly that these loss functions push each token in the hypothesis sequence toward the average of the tokens in the reference, resulting in a poor training signal., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Economic uncertainty: a geometric indicator of discrepancy among experts’ expectations

  • Claveria González, Oscar
  • Monte Moreno, Enrique|||0000-0002-4907-0494
  • Torra Porras, Salvador
The final publication is available at Springer via http://dx.doi.org/10.1007/s11205-018-1984-2, In this study we present a geometric approach to proxy economic uncertainty. We design a positional indicator of disagreement among survey-based agents’ expectations about the state of the economy. Previous dispersion-based uncertainty indicators derived from business and consumer surveys exclusively make use of the two extreme pieces of information: the percentage of respondents expecting a variable to rise and to fall. With the aim of also incorporating the information coming from the share of respondents expecting a variable to remain constant, we propose a geometrical framework and use a barycentric coordinate system to generate a measure of disagreement, referred to as a discrepancy indicator. We assess its performance both empirically and experimentally by comparing it to the standard deviation of the share of positive and negative responses. When applied in sixteen European countries, we find that both time-varying metrics co-evolve in most countries for expectations about the country’s overall economic situation in the present, but not in the future. Additionally, we obtain their simulated sampling distributions and we find that the proposed indicator gravitates uniformly towards the three vertices of the simplex representing the three answering categories, as opposed to the standard deviation, which tends to overestimate the level of uncertainty as a result of ignoring the no-change responses. Consequently, we find evidence that the information coming from agents expecting a variable to remain constant has an effect on the measurement of disagreement., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Unemployment expectations: A socio-demographic analysis of the effect of news

  • Soric, Petar
  • Lolic, Ivana
  • Claveria González, Oscar
  • Monte Moreno, Enrique|||0000-0002-4907-0494
  • Torra Porras, Salvador
© <2019>. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/, In this study, we evaluate the effect of news on consumer unemployment expectations for sixteen socio-demographic groups. To this end, we construct an unemployment sentiment indicator and extract news about several economic variables. By means of genetic programming we estimate symbolic regressions that link unemployment rates in the Euro Area to qualitative expectations about a wide range of economic variables. We then use the evolved expressions to compute unemployment expectations for each consumer group. We first assess the out-of-sample forecast accuracy of the evolved indicators, obtaining better forecasts for the leading unemployment sentiment indicator than for the coincident one. Results are similar across the different socio-demographic groups. The best forecast results are obtained for respondents between 30 and 49 years. The group where we observe the bigger differences among categories is the occupation, where the lowest forecast errors are obtained for the unemployed respondents. Next, we link news about inflation, industrial production, and stock markets to unemployment expectations. With this aim we match positive and negative news with consumers’ unemployment sentiment using a distributed lag regression model for each news item. We find asymmetries in the responses of consumers’ unemployment expectations to economic news: they tend to be stronger in the case of negative news, especially in the case of inflation., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

UPC multimodal speaker diarization system for the 2018 Albayzin challenge

  • India Massana, Miquel Àngel|||0000-0002-3107-3662
  • Sagastiberri, Itziar
  • Palau Puigdevall, Ponç
  • Sayrol Clols, Elisa|||0000-0002-0526-9733
  • Morros Rubió, Josep Ramon|||0000-0002-1395-487X
  • Hernando Pericás, Francisco Javier|||0000-0002-1730-8154
This paper presents the UPC system proposed for the Multimodal Speaker Diarization task of the 2018 Albayzin Challenge. This approach works by processing individually the speech and the image signal. In the speech domain, speaker diarization is performed using identity embeddings created by a triplet loss DNN that uses i-vectors as input. The triplet DNN is trained with an additional regularization loss that minimizes the variance of both positive and negative distances. A sliding windows is then used to compare speech segments with enrollment speaker targets using cosine distance between the embeddings. To detect identities from the face modality, a face detector followed by a face tracker has been used on the videos. For each cropped face a feature vector is obtained using a Deep Neural Network based on the ResNet 34 architecture, trained using a metric learning triplet loss (available from dlib library). For each track the face feature vector is obtained by averaging the features obtained for each one of the frames of that track. Then, this feature vector is compared with the features extracted from the images of the enrollment identities. The proposed system is evaluated on the RTVE2018 database., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Estudio, informe, memoria (Report)

Guidelines for producing a database of continuous acoustic environment recordings and child physiological variables in a neonatal intensive care unit

  • Nadeu Camprubí, Climent|||0000-0002-5863-0983
  • Raboshchuk, Ganna
Internal document produced in a collaborative project with Hospital Materno-Infantil Sant Joan de Déu de Barcelona.
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Refinement of unsupervised cross-lingual word embeddings

  • Biesialska, Magdalena Marta|||0000-0001-7890-3523
  • Ruiz Costa-Jussà, Marta|||0000-0002-5703-520X
Cross-lingual word embeddings aim to bridge the gap between high-resource and low-resource languages by allowing to learn multilingual word representations even without using any direct bilingual signal. The lion's share of the methods are projection-based approaches that map pre-trained embeddings into a shared latent space. These methods are mostly based on the orthogonal transformation, which assumes language vector spaces to be isomorphic. However, this criterion does not necessarily hold, especially for morphologically-rich languages. In this paper, we propose a self-supervised method to refine the alignment of unsupervised bilingual word embeddings. The proposed model moves vectors of words and their corresponding translations closer to each other as well as enforces length- and center-invariance, thus allowing to better align cross-lingual embeddings. The experimental results demonstrate the effectiveness of our approach, as in most cases it outperforms state-of-the-art methods in a bilingual lexicon induction task., We thank anonymous reviewers for their helpful comments. This work is supported in part by the Spanish Ministerio de Economía y Competitividad, the European Regional Development Fund and the Agencia Estatal de Investigación, through the post-doctoral senior grant Ramón y Cajal, the contract TEC2015-69266-P (MINECO/FEDER,EU) and the contract PCIN-2017-079 (AEI/MINECO)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Using survey data to forecast real activity with evolutionary algorithms. A cross-country analysis

  • Claveria, Oscar
  • Monte Moreno, Enrique|||0000-0002-4907-0494
  • Torra Porras, Salvador
In this study we use survey expectations about a wide range of economic variables to forecast real activity. We propose an empirical approach to derive mathematical functional forms that link survey expectations to economic growth. Combining symbolic regression with genetic programming we generate two survey-based indicators: a perceptions index, using agents’assessments about the present, and an expectations index with their expectations about the future. In order to find the optimal combination of both indexes that best replicates the evolution of economic activity in each country we use a portfolio management procedure known as index tracking. By means of a generalized reduced gradient algorithm we derive the relative weights of both indexes. In most economies, the survey-based predictions generated with the composite indicator outperform the benchmark model for one-quarter ahead forecasts, although these improvements are only significant in Austria, Belgium and Portugal., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Time-domain speech enhancement using generative adversarial networks

  • Pascual de la Puente, Santiago|||0000-0002-8365-7387
  • Serra, Joan
  • Bonafonte Cávez, Antonio|||0000-0002-6240-9915
Speech enhancement improves recorded voice utterances to eliminate noise that might be impeding their intelligibility or compromising their quality. Typical speech enhancement systems are based on regression approaches that subtract noise or predict clean signals. Most of them do not operate directly on waveforms. In this work, we propose a generative approach to regenerate corrupted signals into a clean version by using generative adversarial networks on the raw signal. We also explore several variations of the proposed system, obtaining insights into proper architectural choices for an adversarially trained, convolutional autoencoder applied to speech. We conduct both objective and subjective evaluations to assess the performance of the proposed method. The former helps us choose among variations and better tune hyperparameters, while the latter is used in a listening experiment with 42 subjects, confirming the effectiveness of the approach in the real world. We also demonstrate the applicability of the approach for more generalized speech enhancement, where we have to regenerate voices from whispered signals., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Multi-output RNN-LSTM for multiple speaker speech synthesis with a-interpolation model

  • Pascual, Santiago
  • Bonafonte Cávez, Antonio|||0000-0002-6240-9915
Deep Learning has been applied successfully to speech processing. In this paper we propose an architecture for speech synthesis using multiple speakers. Some hidden layers are shared by all the speakers, while there is a specific output layer for each speaker. Objective and perceptual experiments prove that this scheme produces much better results in comparison with sin- gle speaker model. Moreover, we also tackle the problem of speaker interpolation by adding a new output layer (a-layer) on top of the multi-output branches. An identifying code is injected into the layer together with acoustic features of many speakers. Experiments show that the a-layer can effectively learn to interpolate the acoustic features between speakers., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Towards mitigating gender bias in a decoder-based neural machine translation model by adding contextual information

  • Basta, Christine Raouf Saad
  • Ruiz Costa-Jussà, Marta|||0000-0002-5703-520X
  • Rodríguez Fonollosa, José Adrián|||0000-0001-9513-7939
Gender bias negatively impacts many natural language processing applications, including ma-chine translation (MT). The motivation behind this work is to study whether recent proposedMT techniques are significantly contributing to attenuate biases in document-level and gender-balanced data. For the study, we consider approaches of adding the previous sentence and thespeaker information, implemented in a decoder-based neural MT system. We show improve-ments both in translation quality (+1 BLEU point) as well as in gender bias mitigation onWinoMT (+5% accuracy)., This work is supported in part by the Catalan Agency for Management of University and Research Grants (AGAUR) through the FI PhD Scholarship. This work is also supported in part by the Spanish Ministerio de Economía y Competitividad, the European Regional Development Fund and the Agencia Estatal de Investigacion, through the postdoctoral senior grant Ramón y Cajal, contract TEC2015-69266-P (MINECO/FEDER,EU) and contract PCIN-2017-079 (AEI/MINECO)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Time series features and machine learning forecasts

  • Claveria González, Oscar
  • Monte Moreno, Enrique|||0000-0002-4907-0494
  • Torra Porras, Salvador
In this study we combine the results of two independent analyses to position Spanish regions according to both the characteristics of the time series of international tourist arrivals and the accuracy of predictions of arrivals at the regional level. We apply a seasonal-trend decomposition procedure based on non-parametric regression to isolate the different components of the series and calculate the main time series features. Predictions are generated with several machine learning models in a recursive multi-step-ahead forecasting experiment. Finally, we summarize all the information from the two previous experiments using categorical principal component analysis. By overlapping the distribution of the regions and the component loadings of each variable along both dimensions, we observe that entropy and dispersion show an inverse relation with forecast accuracy, but the interactions between the rest of the features and accuracy are heavily dependent on the forecast horizon. On this evidence, we conclude that in order to increase forecast accuracy of tourist arrivals at the regional level, model selection should be region-specific and based on the forecast horizon., This research was supported by by the projects ECO2016-75805-R and TEC2015-69266-P from the Spanish Ministry of Economy and Competitiveness. We would like to thank the Editor and two anonymous referees for their useful comments and suggestions., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

The TALP-UPC machine translation systems for WMT19 news translation task: pivoting techniques for low resource MT

  • Casas Manzanares, Noé
  • Rodríguez Fonollosa, José Adrián|||0000-0001-9513-7939
  • Escolano Peinado, Carlos|||0000-0001-6657-673X
  • Basta, Christine Raouf Saad
  • Ruiz Costa-Jussà, Marta|||0000-0002-5703-520X
In this article, we describe the TALP-UPC research group participation in the WMT19 news translation shared task for Kazakh-English. Given the low amount of parallel training data, we resort to using Russian as pivot language, training subword-based statistical translation systems for Russian-Kazakh and Russian-English that were then used to create two synthetic pseudo-parallel corpora for Kazakh-English and English-Kazakh respectively. Finally, a self-attention model based on the decoder part of the Transformer architecture was trained on the two pseudo-parallel corpora., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Unsupervised training of siamese networks for speaker verification

  • Khan, Umair
  • Hernando Pericás, Francisco Javier|||0000-0002-1730-8154
Speaker labeled background data is an essential requirement for most state-of-the-art approaches in speaker recognition, e.g., xvectors and i-vector/PLDA. However, in reality it is difficult to access large amount of labeled data. In this work, we propose siamese networks for speaker verification without using speaker labels. We propose two different siamese networks having two and three branches, respectively, where each branch is a CNN encoder. Since the goal is to avoid speaker labels, we propose to generate the training pairs in an unsupervised manner. The client samples are selected within one database according to highest cosine scores with the anchor in i-vector space. The impostor samples are selected in the same way but from another database. Our double-branch siamese performs binary classification using cross entropy loss during training. In testing phase, we obtain speaker verification scores directly from its output layer. Whereas, our triple-branch siamese is trained to learn speaker embeddings using triplet loss. During testing, we extract speaker embeddings from its output layer, which are scored in the experiments using cosine scoring. The evaluation is performed on VoxCeleb-1 database, which show that using the proposed unsupervised systems, solely or in fusion, the results get closer to supervised baseline, This work has been developed in the framework of DeepVoice Project (TEC2015-69266-P), funded by Spanish Ministry, Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Wav2Pix: speech-conditioned face generation using generative adversarial networks

  • Cardoso Duarte, Amanda
  • Roldan, Francisco
  • Tubau, Miquel
  • Escur, Janna
  • Pascual de la Puente, Santiago|||0000-0002-8365-7387
  • Salvador Aguilera, Amaia|||0000-0002-9908-1685
  • Mohedano, Eva
  • McGuinness, Kevin
  • Torres Viñals, Jordi|||0000-0003-1963-7418
  • Giró Nieto, Xavier|||0000-0002-9935-5332
Speech is a rich biometric signal that contains information about the identity, gender and emotional state of the speaker. In this work, we explore its potential to generate face images of a speaker by conditioning a Generative Adversarial Network (GAN) with raw speech input. We propose a deep neural network that is trained from scratch in an end-to-end fashion, generating a face directly from the raw speech waveform without any additional identity information (e.g reference image or one-hot encoding). Our model is trained in a self-supervised approach by exploiting the audio and visual signals naturally aligned in videos. With the purpose of training from video data, we present a novel dataset collected for this work, with high-quality videos of youtubers with notable expressiveness in both the speech and visual signals., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Estudio, informe, memoria (Report)

Let the data do the talking: empirical modelling of survey-based expectations by means of genetic programming

  • Claveria González, Oscar
  • Monte Moreno, Enrique|||0000-0002-4907-0494
  • Torra Porras, Salvador
Working paper, In this study we use agents’ expectations about the state of the economy to generate indicators of economic activity in twenty-six European countries grouped in five regions (Western, Eastern, and Southern Europe, and Baltic and Scandinavian countries). We apply a data-driven procedure based on evolutionary computation to transform survey variables in economic growth rates. In a first step, we design five independent experiments to derive the optimal combination of expectations that best replicates the evolution of economic growth in each region by means of genetic programming, limiting the integration schemes to the main mathematical operations. We then rank survey variables according to their performance in tracking economic activity, finding that agents’ “perception about the overall economy compared to last year” is the survey variable with the highest predictive power. In a second step, we assess the out-of-sample forecast accuracy of the evolved indicators. Although we obtain different results across regions, Austria, Slovakia, Portugal, Lithuania and Sweden are the economies of each region that show the best forecast results. We also find evidence that the forecasting performance of the survey-based indicators improves during periods of higher growth.
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Multi-speaker neural vocoder

  • Barbany Mayor, Oriol|||0000-0002-1379-9152
  • Bonafonte Cávez, Antonio|||0000-0002-6240-9915
  • Pascual de la Puente, Santiago|||0000-0002-8365-7387
Statistical Parametric Speech Synthesis (SPSS) offers more f lexibility than unit-selection based speech synthesis, which was the dominant commercial technology during the 2000s decade. However, classical SPSS systems generate speech with lower naturalness than unit-selection methods. Deep learning based SPSS, thanks to recurrent architectures, surpasses classical SPSS limits. These architectures offer high quality speech while preserving the desired flexibility in choosing the parameters such as the speaker, the intonation, etc. This paper exposes two proposals conceived to improve deep learning-based text-to-speech systems. First a baseline model, obtained by adapting SampleRNN, making it as a speaker-independent neural vocoder that generates the speech waveform from acoustic parameters. Then two approaches are proposed to improve the quality, applying speaker dependent normalization of the acoustic features, and the look ahead, consisting on feeding acoustic features of future frames to the network with the aim of better modeling the present waveform and avoiding possible discontinuities. Human listeners prefer the system that combines both techniques, which reaches a rate of 4 in the mean opinion score scale (MOS) with the balanced dataset and outperforms the other models., This research was supported by the project TEC2015-69266-P (MINECO/FEDER, UE)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

The TALP-UPC system for the WMT similar language task: statistical vs neural machine translation

  • Biesialska, Magdalena Marta|||0000-0001-7890-3523
  • Guàrdia Fernández, Lluís
  • Ruiz Costa-Jussà, Marta|||0000-0002-5703-520X
Although the problem of similar language translation has been an area of research interest for many years, yet it is still far from being solved. In this paper, we study the performance of two popular approaches: statistical and neural. We conclude that both methods yield similar results; however, the performance varies depending on the language pair. While the statistical approach outperforms the neural one by a difference of 6 BLEU points for the Spanish-Portuguese language pair, the proposed neural model surpasses the statistical one by a difference of 2 BLEU points for Czech-Polish. In the former case, the language similarity (based on perplexity) is much higher than in the latter case. Additionally, we report negative results for the system combination with back-translation. Our TALP-UPC system submission won 1st place for Czech->Polish and 2nd place for Spanish->Portuguese in the official evaluation of the 1st WMT Similar Language Translation task., This work is supported in part by the Span-ish Ministerio de Economía y Competitividad,the European Regional Development Fund andthe Agencia Estatal de Investigación, through thepostdoctoral senior grant Ramón y Cajal, the con-tract TEC2015-69266-P (MINECO/FEDER,EU)and the contract PCIN-2017-079 (AEI/MINECO)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Linguistic knowledge-based vocabularies for Neural Machine Translation

  • Casas Manzanares, Noé
  • Ruiz Costa-Jussà, Marta|||0000-0002-5703-520X
  • Rodríguez Fonollosa, José Adrián|||0000-0001-9513-7939
  • Alonso, Juan
  • Fanlo, Ramon
This article has been published in a revised form in Natural Language Engineering https://doi.org/10.1017/S1351324920000364. This version is free to view and download for private research and study only. Not for re-distribution, re-sale or use in derivative works. © Cambridge University Press, Neural Networks applied to Machine Translation need a finite vocabulary to express textual information as a sequence of discrete tokens. The currently dominant subword vocabularies exploit statistically-discovered common parts of words to achieve the flexibility of character-based vocabularies without delegating the whole learning of word formation to the neural network. However, they trade this for the inability to apply word-level token associations, which limits their use in semantically-rich areas and prevents some transfer learning approaches e.g. cross-lingual pretrained embeddings, and reduces their interpretability. In this work, we propose new hybrid linguistically-grounded vocabulary definition strategies that keep both the advantages of subword vocabularies and the word-level associations, enabling neural networks to profit from the derived benefits. We test the proposed approaches in both morphologically rich and poor languages, showing that, for the former, the quality in the translation of out-of-domain texts is improved with respect to a strong subword baseline., This work is partially supported by Lucy Software / United Language Group (ULG) and the Catalan Agency for Management of University and Research Grants (AGAUR) through an Industrial PhD Grant. This work is also supported in part by the Spanish Ministerio de Economa y Competitividad, the European Regional Development Fund and the Agencia Estatal de Investigacin, through the postdoctoral senior grant Ramn y Cajal, contract TEC2015-69266-P (MINECO/FEDER,EU) and contract PCIN-2017-079 (AEI/MINECO)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Estudio, informe, memoria (Report)

Joint model-based recognition and localization of overlapped acoustic events using a set of distributed small microphone arrays

  • Chakraborty, Rupayan
  • Nadeu Camprubí, Climent|||0000-0002-5863-0983
Article publicat a Arxiv, sense revisió per parells, In the analysis of acoustic scenes, often the occurring sounds have to be detected in time, recognized, and localized in space. Usually, each of these tasks is done separately. In this paper, a model-based approach to jointly carry them out for the case of multiple simultaneous sources is presented and tested. The recognized event classes and their respective room positions are obtained with a single system that maximizes the combination of a large set of scores, each one resulting from a different acoustic event model and a different beamformer output signal, which comes from one of several arbitrarily-located small microphone arrays. By using a two-step method, the experimental work for a specific scenario consisting of meeting-room acoustic events, either isolated or overlapped with speech, is reported. Tests carried out with two datasets show the advantage of the proposed approach with respect to some usual techniques, and that the inclusion of estimated priors brings a further performance improvement.
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Machine and deep learning approaches to localization and range estimation of underwater acoustic sources

  • Houégnigan, Ludwig|||0000-0003-2903-3140
  • Safari, Pooyan
  • Nadeu Camprubí, Climent|||0000-0002-5863-0983
  • André, Michel|||0000-0002-0091-7279
  • Van der Schaar, Mike Connor Roger Malcolm
This paper introduces ongoing experiments and early results for the underwater localization and range estimation of acoustic sources. Beyond classical results obtained for direction of arrival estimation, results concerning range estimation using supervised learning with neural networks having both shallow and deep architectures are presented. The developed method is applicable in the context of a single sensor, a compact array, or a small aperture towed array and provided results with great potential both for industrial impact and for the conservation and density estimation of cetaceans. With an average error of 4.3% and 3.5%-respectively for a shallow and for a deep pre-trained architecture-for ranges up to 8 kilometers and consistently below 300 meters, the system provides robust estimates suitable for an automated real-time solution., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publication

Guidelines for producing a database of continuous acoustic environment recordings and child physiological variables in a neonatal intensive care unit

  • Nadeu Camprubí, Climent
  • Raboshchuk, Ganna
Internal document produced in a collaborative project with Hospital Materno-Infantil Sant Joan de Déu de Barcelona., Preprint
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Machine and deep learning approaches to localization and range estimation of underwater acoustic sources

  • Houégnigan, Ludwig
  • Safari, Pooyan
  • Nadeu Camprubí, Climent
  • André, Michel
  • Van der Schaar, Mike Connor Roger Malcolm
This paper introduces ongoing experiments and early results for the underwater localization and range estimation of acoustic sources. Beyond classical results obtained for direction of arrival estimation, results concerning range estimation using supervised learning with neural networks having both shallow and deep architectures are presented. The developed method is applicable in the context of a single sensor, a compact array, or a small aperture towed array and provided results with great potential both for industrial impact and for the conservation and density estimation of cetaceans. With an average error of 4.3% and 3.5%-respectively for a shallow and for a deep pre-trained architecture-for ranges up to 8 kilometers and consistently below 300 meters, the system provides robust estimates suitable for an automated real-time solution., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Unemployment expectations: A socio-demographic analysis of the effect of news

  • Soric, Petar
  • Lolic, Ivana
  • Claveria González, Oscar
  • Monte Moreno, Enrique
  • Torra Porras, Salvador
© <2019>. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/, In this study, we evaluate the effect of news on consumer unemployment expectations for sixteen socio-demographic groups. To this end, we construct an unemployment sentiment indicator and extract news about several economic variables. By means of genetic programming we estimate symbolic regressions that link unemployment rates in the Euro Area to qualitative expectations about a wide range of economic variables. We then use the evolved expressions to compute unemployment expectations for each consumer group. We first assess the out-of-sample forecast accuracy of the evolved indicators, obtaining better forecasts for the leading unemployment sentiment indicator than for the coincident one. Results are similar across the different socio-demographic groups. The best forecast results are obtained for respondents between 30 and 49 years. The group where we observe the bigger differences among categories is the occupation, where the lowest forecast errors are obtained for the unemployed respondents. Next, we link news about inflation, industrial production, and stock markets to unemployment expectations. With this aim we match positive and negative news with consumers’ unemployment sentiment using a distributed lag regression model for each news item. We find asymmetries in the responses of consumers’ unemployment expectations to economic news: they tend to be stronger in the case of negative news, especially in the case of inflation., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication

A geometric approach to proxy economic uncertainty by a metric of disagreement among qualitative expectations

  • Claveria González, Oscar
  • Monte Moreno, Enrique
  • Torra Porras, Salvador
Working paper, In this study we present a geometric approach to proxy economic uncertainty. We design a positional indicator of disagreement among survey-based agents' expectations about the state of the economy. Previous dispersion-based uncertainty indicators derived from business and consumer surveys exclusively make use of the two extreme pieces of information coming from the respondents expecting a variable to rise and to fall. With the aim of also incorporating the information coming from the share of respondents expecting a variable to remain constant, we propose a geometrical framework and use a barycentric coordinate system to generate a measure of disagreement, referred to as a discrepancy indicator. We assess its performance, both empirically and experimentally, by comparing it to the standard deviation of the share of positive and negative responses, which has been used by Bachman et al. (2013) as a proxy for economic uncertainty. When applied in sixteen European countries, we find that both time-varying metrics co-evolve in most countries for expectations about the country's overall economic situation in the present, but not in the future. Additionally, we obtain their simulated sampling distributions and we find that the proposed indicator gravitates uniformly towards the three vertices of the simplex representing the three answering categories, as opposed to the standard deviation, which tends to overestimate the level of uncertainty as a result of ignoring the no-change responses. Consequently, we find evidence that the information coming from agents expecting a variable to remain constant has an effect on the measurement of disagreement., Preprint
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Prosodic and spectral iVectors for expressive speech synthesis

  • Jauk, Igor
  • Bonafonte Cávez, Antonio
This work presents a study on the suitability of prosodic andacoustic features, with a special focus on i-vectors, in expressivespeech analysis and synthesis. For each utterance of two dif-ferent databases, a laboratory recorded emotional acted speech,and an audiobook, several prosodic and acoustic features are ex-tracted. Among them, i-vectors are built not only on the MFCCbase, but also on F0, power and syllable durations. Then, un-supervised clustering is performed using different feature com-binations. The resulting clusters are evaluated calculating clus-ter entropy for labeled portions of the databases. Additionally,synthetic voices are trained, applying speaker adaptive training,from the clusters built from the audiobook. The voices are eval-uated in a perceptual test where the participants have to edit anaudiobook paragraph using the synthetic voices.The objective results suggest that i-vectors are very use-ful for the audiobook, where different speakers (book charac-ters) are imitated. On the other hand, for the laboratory record-ings, traditional prosodic features outperform i-vectors. Also,a closer analysis of the created clusters suggest that differentspeakers use different prosodic and acoustic means to conveyemotions. The perceptual results suggest that the proposed i-vector based feature combinations can be used for audiobookclustering and voice training., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

A novel approach to real-time range estimation of underwater acoustic sources using supervised machine learning

  • Houégnigan, Ludwig
  • Safari, Pooyan
  • Nadeu Camprubí, Climent
  • Van der Schaar, Mike Connor Roger Malcolm
  • André, Michel
© 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., The proposed paper introduces a novel method for range estimation of acoustic sources, both cetaceans and industrial sources, in deep sea environments using supervised learning with neural networks in the contex of a single sensor, a compact array, or a small aperture towed array. The presented results have potential both for industrial impact and for the conservation and density estimation of cetaceans. With an average error of 4.3% for ranges up to 8 kilometers and typically below 300 meters, those results are challenging and to our knowledge they are unprecedented for an automated real-time solution., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Detection and description of the different ionospheric disturbances that appeared during the solar eclipse of 21 August 2017

  • Yang, Heng
  • Monte Moreno, Enrique
  • Hernández Pajares, Manuel
This work will provide a detailed characterization of the travelling ionospheric disturbances (TIDs) created by the solar eclipse of 21 August 2017, the shadow of which crossed the United States from the Pacific to the Atlantic ocean. The analysis is done by means of the Atomic Decomposition Detector of Traveling Ionospheric Disturbances (ADDTID) algorithm. This method automatically detects and characterizes multiple TIDs from the global navigation satellite system (GNSS) observation. The set of disturbances generated by the eclipse has a richer and more varied behavior than that associated with the shock wave directly produced by cooling effects of the moon shadow. This can be modeled in part as if the umbra and penumbra of the eclipse were moving cylinders that intersects with variable elevation angle a curved surface. This projection gives rise to regions of equal penumbra with shapes similar to ellipses, with different centers and foci. The result of this is reflected in the time evolution of the TID wavelengths produced by the eclipse, which depend on the vertical angle of the sun with the surface of the earth, and also a double bow wave phenomenon, where the bow waves are generated in advance to the umbra. We show that the delay in the appearance of the disturbances with the transit of the eclipse are compatible with the physical explanations, linked to the different origins of the disturbances and the wavelengths. Finally, we detected a consistent pattern, in location and time of disturbances in advance to the penumbra as a set of medium scale TIDs, which could be hypothesized as soliton waves of the bow wave. In all cases, the detected disturbances were checked visually on the detrended vertical total electron content (TEC) maps., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Problem-agnostic speech embeddings for multi-speaker text-to-speech with SampleRNN

  • Álvarez Robert, David
  • Pascual de la Puente, Santiago
  • Bonafonte Cávez, Antonio
Text-to-speech (TTS) acoustic models map linguistic features into an acoustic representation out of which an audible waveform is generated. The latest and most natural TTS systems build a direct mapping between linguistic and waveform domains, like SampleRNN. This way, possible signal naturalness losses are avoided as intermediate acoustic representations are discarded. Another important dimension of study apart from naturalness is their adaptability to generate voice from new speakers that were unseen during training. In this paper we first propose the use of problem-agnostic speech embeddings in a multi-speaker acoustic model for TTS based on SampleRNN. This way, we feed the acoustic model with speaker acousticallydependent representations that enrich the waveform generation more than embeddings unrelated to these factors. Our first results suggest that the proposed embeddings lead to better quality voices than those obtained with one-hot embeddings. Furthermore, as we can use any speech segment as an encoded representation during inference, the model is capable to generalize to new speaker identities without retraining the network. We finally show that, with a small increase of speech duration in the embedding extractor, we dramatically reduce the spectral distortion to close the gap towards the target identities., This research was supported by the project TEC2015-69266-P (MINECO/FEDER, UE)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Wav2Pix: speech-conditioned face generation using generative adversarial networks

  • Cardoso Duarte, Amanda
  • Roldan, Francisco
  • Tubau, Miquel
  • Escur, Janna
  • Pascual de la Puente, Santiago
  • Salvador Aguilera, Amaia
  • Mohedano, Eva
  • McGuinness, Kevin
  • Torres Viñals, Jordi
  • Giró Nieto, Xavier
Speech is a rich biometric signal that contains information about the identity, gender and emotional state of the speaker. In this work, we explore its potential to generate face images of a speaker by conditioning a Generative Adversarial Network (GAN) with raw speech input. We propose a deep neural network that is trained from scratch in an end-to-end fashion, generating a face directly from the raw speech waveform without any additional identity information (e.g reference image or one-hot encoding). Our model is trained in a self-supervised approach by exploiting the audio and visual signals naturally aligned in videos. With the purpose of training from video data, we present a novel dataset collected for this work, with high-quality videos of youtubers with notable expressiveness in both the speech and visual signals., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Multi-speaker neural vocoder

  • Barbany Mayor, Oriol
  • Bonafonte Cávez, Antonio
  • Pascual de la Puente, Santiago
Statistical Parametric Speech Synthesis (SPSS) offers more f lexibility than unit-selection based speech synthesis, which was the dominant commercial technology during the 2000s decade. However, classical SPSS systems generate speech with lower naturalness than unit-selection methods. Deep learning based SPSS, thanks to recurrent architectures, surpasses classical SPSS limits. These architectures offer high quality speech while preserving the desired flexibility in choosing the parameters such as the speaker, the intonation, etc. This paper exposes two proposals conceived to improve deep learning-based text-to-speech systems. First a baseline model, obtained by adapting SampleRNN, making it as a speaker-independent neural vocoder that generates the speech waveform from acoustic parameters. Then two approaches are proposed to improve the quality, applying speaker dependent normalization of the acoustic features, and the look ahead, consisting on feeding acoustic features of future frames to the network with the aim of better modeling the present waveform and avoiding possible discontinuities. Human listeners prefer the system that combines both techniques, which reaches a rate of 4 in the mean opinion score scale (MOS) with the balanced dataset and outperforms the other models., This research was supported by the project TEC2015-69266-P (MINECO/FEDER, UE)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

The TALP-UPC machine translation systems for WMT19 news translation task: pivoting techniques for low resource MT

  • Casas Manzanares, Noé
  • Rodríguez Fonollosa, José Adrián
  • Escolano Peinado, Carlos
  • Basta, Christine Raouf Saad
  • Ruiz Costa-Jussà, Marta
In this article, we describe the TALP-UPC research group participation in the WMT19 news translation shared task for Kazakh-English. Given the low amount of parallel training data, we resort to using Russian as pivot language, training subword-based statistical translation systems for Russian-Kazakh and Russian-English that were then used to create two synthetic pseudo-parallel corpora for Kazakh-English and English-Kazakh respectively. Finally, a self-attention model based on the decoder part of the Transformer architecture was trained on the two pseudo-parallel corpora., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Using survey data to forecast real activity with evolutionary algorithms. A cross-country analysis

  • Claveria, Oscar
  • Monte Moreno, Enrique
  • Torra Porras, Salvador
In this study we use survey expectations about a wide range of economic variables to forecast real activity. We propose an empirical approach to derive mathematical functional forms that link survey expectations to economic growth. Combining symbolic regression with genetic programming we generate two survey-based indicators: a perceptions index, using agents’assessments about the present, and an expectations index with their expectations about the future. In order to find the optimal combination of both indexes that best replicates the evolution of economic activity in each country we use a portfolio management procedure known as index tracking. By means of a generalized reduced gradient algorithm we derive the relative weights of both indexes. In most economies, the survey-based predictions generated with the composite indicator outperform the benchmark model for one-quarter ahead forecasts, although these improvements are only significant in Austria, Belgium and Portugal., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication

A regional perspective on the accuracy of machine learning forecasts of tourism demand based on data characteristics

  • Claveria González, Oscar
  • Monte Moreno, Enrique
  • Torra Porras, Salvador
Working paper, In this work we assess the role of data characteristics in the accuracy of machine learning (ML) tourism forecasts from a spatial perspective. First, we apply a seasonal-trend decomposition procedure based on non-parametric regression to isolate the different components of the time series of international tourism demand to all Spanish regions. This approach allows us to compute a set of measures to describe the features of the data. Second, we analyse the performance of several ML models in a recursive multiple-step-ahead forecasting experiment. In a third step, we rank all seventeen regions according to their characteristics and the obtained forecasting performance, and use the rankings as the input for a multivariate analysis to evaluate the interactions between time series features and the accuracy of the predictions. By means of dimensionality reduction techniques we summarise all the information into two components and project all Spanish regions into perceptual maps. We find that entropy and dispersion show a negative relation with accuracy, while the effect of other data characteristics on forecast accuracy is heavily dependent on the forecast horizon., Preprint
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Self-attention encoding and pooling for speaker recognition

  • Safari, Pooyan
  • India Massana, Miquel Àngel
  • Hernando Pericás, Francisco Javier
The computing power of mobile devices limits the end-user applications in terms of storage size, processing, memory and energy consumption. These limitations motivate researchers for the design of more efficient deep models. On the other hand, self-attention networks based on Transformer architecture have attracted remarkable interests due to their high parallelization capabilities and strong performance on a variety of Natural Language Processing (NLP) applications. Inspired by the Transformer, we propose a tandem Self-Attention Encoding and Pooling (SAEP) mechanism to obtain a discriminative speaker embedding given non-fixed length speech utterances. SAEP is a stack of identical blocks solely relied on self-attention and position-wise feed-forward networks to create vector representation of speakers. This approach encodes short-term speaker spectral features into speaker embeddings to be used in text-independent speaker verification. We have evaluated this approach on both VoxCeleb1 & 2 datasets. The proposed architecture is able to outperform the baseline x-vector, and shows competitive performance to some other benchmarks based on convolutions, with a significant reduction in model size. It employs 94%, 95%, and 73% less parameters compared to ResNet-34, ResNet-50, and x-vector, respectively. This indicates that the proposed fully attention based architecture is more efficient in extracting time-invariant features from speaker utterances., This work was supported in part by the Spanish Project DeepVoice (TEC2015-69266-P)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

UPC multimodal speaker diarization system for the 2018 Albayzin challenge

  • India Massana, Miquel Àngel
  • Sagastiberri, Itziar
  • Palau Puigdevall, Ponç
  • Sayrol Clols, Elisa
  • Morros Rubió, Josep Ramon
  • Hernando Pericás, Francisco Javier
This paper presents the UPC system proposed for the Multimodal Speaker Diarization task of the 2018 Albayzin Challenge. This approach works by processing individually the speech and the image signal. In the speech domain, speaker diarization is performed using identity embeddings created by a triplet loss DNN that uses i-vectors as input. The triplet DNN is trained with an additional regularization loss that minimizes the variance of both positive and negative distances. A sliding windows is then used to compare speech segments with enrollment speaker targets using cosine distance between the embeddings. To detect identities from the face modality, a face detector followed by a face tracker has been used on the videos. For each cropped face a feature vector is obtained using a Deep Neural Network based on the ResNet 34 architecture, trained using a metric learning triplet loss (available from dlib library). For each track the face feature vector is obtained by averaging the features obtained for each one of the frames of that track. Then, this feature vector is compared with the features extracted from the images of the enrollment identities. The proposed system is evaluated on the RTVE2018 database., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Restricted Boltzmann Machine vectors for speaker clustering

  • Khan, Umair
  • Safari, Pooyan
  • Hernando Pericás, Francisco Javier
Restricted Boltzmann Machines (RBMs) have been used both in the front-end and backend of speaker verification systems. In this work, we apply RBMs as a front-end in the context of speaker clustering. Speakers' utterances are transformed into a vector representation by means of RBMs. These vectors, referred to as RBM vectors, have shown to preserve speaker-specific information and are used for the task of speaker clustering. In this work, we perform the traditional bottom-up Agglomerative Hierarchical Clustering (AHC). Using the RBM vector representation of speakers, the performance of speaker clustering is improved. The evaluation has been performed on the audio recordings of Catalan TV Broadcast shows. The experimental results show that our proposed system outperforms the baseline i-vectors system in terms of Equal Impurity (EI). Using cosine scoring, a relative improvement of 11% and 12% are achieved for average and single linkage clustering algorithms respectively. Using PLDA scoring, the RBM vectors achieve a relative improvement of 11% compared to i-vectors for the single linkage algorithm., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

I-vector transformation using k-nearest neighbors for speaker verification

  • Khan, Umair
  • India Massana, Miquel Àngel
  • Hernando Pericás, Francisco Javier
© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. https://dx.doi.org/10.1109/ICASSP40776.2020.9053504, Probabilistic Linear Discriminant Analysis (PLDA) is the most efficient backend for i-vectors. However, it requires labeled background data which can be difficult to access in practice. Unlike PLDA, cosine scoring avoids speaker-labels at the cost of degrading the performance. In this work, we propose a post processing of i-vectors using a Deep Neural Network (DNN) to transform i-vectors into a new speaker vector representation. The DNN will be trained using i-vectors that are similar to the training i-vectors. These similar i-vectors will be selected in an unsupervised manner. Using the new vector representation, we will score the experimental trials using cosine scoring. The evaluation was performed on the speaker verification trials of VoxCeleb-1 database. The experiments have shown that with the help of the similar i-vectors the new vectors become more discriminative than the original i-vectors. The new vectors have gained a relative improvement of 53% in terms of EER, compared to the conventional i-vector/PLDA system, but without using speaker labels., This work has been developed in the framework of DeepVoice Project (TEC2015-69266-P), funded by Spanish Ministry., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication

Automatic detection of alarm sounds in a noisy hospital environment using model and non-model based approaches

  • Raboshchuk, Ganna
  • Gómez Quintana, Sergi
  • Peiró Lilja, Alexandre
  • Nadeu Camprubí, Climent
Article publicat sense revisió per parells a Arxiv, In the noisy acoustic environment of a Neonatal Intensive Care Unit (NICU) there is a variety of alarms, which are frequently triggered by the biomedical equipment. In this paper different approaches for automatic detection of those sound alarms are presented and compared: 1) a non-model-based approach that employs signal processing techniques; 2) a model-based approach based on neural networks; and 3) an approach that combines both non-model and model-based approaches. The performance of the developed detection systems that follow each of those approaches is assessed, analysed and compared both at the frame level and at the event level by using an audio database recorded in a real-world hospital environment., Preprint
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Empirical modelling of survey-based expectations for the design of economic indicators in five European regions

  • Claveria, Oscar
  • Monte Moreno, Enrique
  • Torra Porras, Salvador
This is a post-peer-review, pre-copyedit version of an article published in Empirica. The final authenticated version is available online at: http://dx.doi.org/10.1007/s10663-017-9395-1”., In this study we use agents’ expectations about the state of the economy to generate indicators of economic activity in twenty-six European countries grouped in five regions (Western, Eastern, and Southern Europe, and Baltic and Scandinavian countries). We apply a data-driven procedure based on evolutionary computation to transform survey variables in economic growth rates. In a first step, we design five independent experiments to derive a formula using survey variables that best replicates the evolution of economic growth in each region by means of genetic programming, limiting the integration schemes to the main mathematical operations. We then rank survey variables according to their performance in tracking economic activity, finding that agents’ ‘‘perception about the overall economy compared to last year’’ is the survey variable with the highest predictive power. In a second step, we assess the out-of-sample forecast accuracy of the evolved indicators. Although we obtain different results across regions, Austria, Slovakia, Portugal, Lithuania and Sweden are the economies of each region that show the best forecast results. We also find evidence that the forecasting performance of the survey-based indicators improves during periods of higher growth., This is a post-peer-review, pre-copyedit version of an article published in Empirica. The final authenticated version is available online at: http://dx.doi.org/10.1007/s10663-017-9395-1”., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

A differentiable BLEU loss. Analysis and first results

  • Casas Manzanares, Noé
  • Rodríguez Fonollosa, José Adrián
  • Ruiz Costa-Jussà, Marta
In natural language generation tasks, like neural machine translation and image captioning, there is usually a mismatch between the optimized loss and the de facto evaluation criterion, namely token-level maximum likelihood and corpus-level BLEU score. This article tries to reduce this gap by defining differentiable computations of the BLEU and GLEU scores. We test this approach on simple tasks, obtaining valuable lessons on its potential applications but also its pitfalls, mainly that these loss functions push each token in the hypothesis sequence toward the average of the tokens in the reference, resulting in a poor training signal., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Linguistic knowledge-based vocabularies for Neural Machine Translation

  • Casas Manzanares, Noé
  • Ruiz Costa-Jussà, Marta
  • Rodríguez Fonollosa, José Adrián
  • Alonso, Juan
  • Fanlo, Ramon
This article has been published in a revised form in Natural Language Engineering https://doi.org/10.1017/S1351324920000364. This version is free to view and download for private research and study only. Not for re-distribution, re-sale or use in derivative works. © Cambridge University Press, Neural Networks applied to Machine Translation need a finite vocabulary to express textual information as a sequence of discrete tokens. The currently dominant subword vocabularies exploit statistically-discovered common parts of words to achieve the flexibility of character-based vocabularies without delegating the whole learning of word formation to the neural network. However, they trade this for the inability to apply word-level token associations, which limits their use in semantically-rich areas and prevents some transfer learning approaches e.g. cross-lingual pretrained embeddings, and reduces their interpretability. In this work, we propose new hybrid linguistically-grounded vocabulary definition strategies that keep both the advantages of subword vocabularies and the word-level associations, enabling neural networks to profit from the derived benefits. We test the proposed approaches in both morphologically rich and poor languages, showing that, for the former, the quality in the translation of out-of-domain texts is improved with respect to a strong subword baseline., This work is partially supported by Lucy Software / United Language Group (ULG) and the Catalan Agency for Management of University and Research Grants (AGAUR) through an Industrial PhD Grant. This work is also supported in part by the Spanish Ministerio de Economa y Competitividad, the European Regional Development Fund and the Agencia Estatal de Investigacin, through the postdoctoral senior grant Ramn y Cajal, contract TEC2015-69266-P (MINECO/FEDER,EU) and contract PCIN-2017-079 (AEI/MINECO)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Byte-based neural machine translation

  • Ruiz Costa-Jussà, Marta
  • Escolano Peinado, Carlos
  • Rodríguez Fonollosa, José Adrián
This paper presents experiments compar- ing character-based and byte-based neural machine translation systems. The main motivation of the byte-based neural ma- chine translation system is to build multi- lingual neural machine translation systems that can share the same vocabulary. We compare the performance of both systems in several language pairs and we see that the performance in test is similar for most language pairs while the training time is slightly reduced in the case of byte-based neural machine translation., Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Time-domain speech enhancement using generative adversarial networks

  • Pascual de la Puente, Santiago
  • Serra, Joan
  • Bonafonte Cávez, Antonio
Speech enhancement improves recorded voice utterances to eliminate noise that might be impeding their intelligibility or compromising their quality. Typical speech enhancement systems are based on regression approaches that subtract noise or predict clean signals. Most of them do not operate directly on waveforms. In this work, we propose a generative approach to regenerate corrupted signals into a clean version by using generative adversarial networks on the raw signal. We also explore several variations of the proposed system, obtaining insights into proper architectural choices for an adversarially trained, convolutional autoencoder applied to speech. We conduct both objective and subjective evaluations to assess the performance of the proposed method. The former helps us choose among variations and better tune hyperparameters, while the latter is used in a listening experiment with 42 subjects, confirming the effectiveness of the approach in the real world. We also demonstrate the applicability of the approach for more generalized speech enhancement, where we have to regenerate voices from whispered signals., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

The TALP-UPC system for the WMT similar language task: statistical vs neural machine translation

  • Biesialska, Magdalena Marta
  • Guàrdia Fernández, Lluís
  • Ruiz Costa-Jussà, Marta
Although the problem of similar language translation has been an area of research interest for many years, yet it is still far from being solved. In this paper, we study the performance of two popular approaches: statistical and neural. We conclude that both methods yield similar results; however, the performance varies depending on the language pair. While the statistical approach outperforms the neural one by a difference of 6 BLEU points for the Spanish-Portuguese language pair, the proposed neural model surpasses the statistical one by a difference of 2 BLEU points for Czech-Polish. In the former case, the language similarity (based on perplexity) is much higher than in the latter case. Additionally, we report negative results for the system combination with back-translation. Our TALP-UPC system submission won 1st place for Czech->Polish and 2nd place for Spanish->Portuguese in the official evaluation of the 1st WMT Similar Language Translation task., This work is supported in part by the Span-ish Ministerio de Economía y Competitividad,the European Regional Development Fund andthe Agencia Estatal de Investigación, through thepostdoctoral senior grant Ramón y Cajal, the con-tract TEC2015-69266-P (MINECO/FEDER,EU)and the contract PCIN-2017-079 (AEI/MINECO)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

A geometric proxy of economic uncertainty based on the disagreement in survey expectations

  • Claveria González, Oscar
  • Monte Moreno, Enrique
  • Torra Porras, Salvador
In this study we present a geometric approach to proxy economic uncertainty. We design a positional indicator of disagreement among survey-based agents' expectations about the state of the economy. Previous dispersion-based uncertainty measures derived from business and consumer surveys exclusively make use of the two extreme pieces of information: the percentage of respondents expecting a variable to rise and to fall. With the aim of also incorporating the information coming from the share of respondents expecting a variable to remain constant, we propose a geometrical framework and use a barycentric coordinate system to generate a metric of disagreement, referred to as a discrepancy indicator. We assess its performance, both empirically and experimentally, by comparing it to the standard deviation of the share of positive and negative responses, which has been used by Bachman et al. (2013) as a proxy for economic uncertainty. When applied in sixteen European countries, we find that both time-varying metrics co-evolve in most countries for expectations about the country's overall economic situation in the present, but not in the future. Additionally, we obtain their simulated sampling distributions and we find that the proposed indicator gravitates uniformly towards the three vertices of the simplex representing the three answering categories, as opposed to the standard deviation, which tends to overestimate the level of uncertainty as a result of ignoring the no-change responses. Consequently, we find evidence that the information coming from agents expecting a variable to remain constant has an effect on the measurement of disagreement., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Multilingual, multi-scale and multi-layer visualization of sequence-based intermediate representations

  • Escolano Peinado, Carlos
  • Ruiz Costa-Jussà, Marta
  • Lacroux, Elora
  • Vázquez Alcocer, Pere Pau
The main alternatives nowadays to dealwith sequences are Recurrent Neural Net-works (RNN), Convolutional Neural Networks(CNN) architectures and the Transformer. Inthis context, RNN’s, CNN’s and Transformerhave most commonly been used as an encoder-decoder architecture with multiple layers ineach module. Far beyond this, these architec-tures are the basis for the contextual word em-beddings which are revolutionizing most natural language downstream applications. However, intermediate layer representations insequence-based architectures can be difficultto interpret. To make each layer representation within these architectures more accessible and meaningful, we introduce a web-based toolthat visualizes them both at the sentence and token level. We present three use cases. The first analyses gender issues in contextual worde mbeddings. The second and third are show-ing multilingual intermediate representations for sentences and tokens and the evolution of these intermediate representations along the multiple layers of the decoder and in the con-text of multilingual machine translation., This work is supported by a Google Faculty Research Award. This workis also supported by the Spanish Ministerio de Economía y Competitividad, the European Regional Development Fund and the Agencia Estatal de Investigación, through the post-doctoral senior grant Ramón y Cajal, contracts TEC2015-69266-P and TIN2017-88515-C2-1-R(GEN3DLIVE) (MINECO/FEDER,EU), and contract PCIN-2017-079 (AEI/MINECO)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

TEC forecasting based on manifold trajectories

  • Monte Moreno, Enrique
  • García Rigo, Alberto
  • Hernández Pajares, Manuel
  • Yang, Heng
In this paper, we present a method for forecasting the ionospheric Total Electron Content (TEC) distribution from the International GNSS Service’s Global Ionospheric Maps. The forecasting system gives an estimation of the value of the TEC distribution based on linear combination of previous TEC maps (i.e., a set of 2D arrays indexed by time), and the computation of a tangent subspace in a manifold associated to each map. The use of the tangent space to each map is justified because it allows modeling the possible distortions from one observation to the next as a trajectory on the tangent manifold of the map. The coefficients of the linear combination of the last observations along with the tangent space are estimated at each time stamp to minimize the mean square forecasting error with a regularization term. The estimation is made at each time stamp to adapt the forecast to short-term variations in solar activity., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Evolutionary computation for macroeconomic forecasting

  • Claveria, Oscar
  • Monte Moreno, Enrique
  • Torra Porras, Salvador
The final publication is available at Springer via http://dx.doi.org/10.1007/s10614-017-9767-4, The main objective of this study is twofold. First, we propose an empirical modelling approach based on genetic programming to forecast economic growth by means of survey data on expectations. We use evolutionary algorithms to estimate a symbolic regression that links survey-based expectations to a quantitative variable used as a yardstick, deriving mathematical functional forms that approximate the target variable. The set of empirically-generated proxies of economic growth are used as building blocks to forecast the evolution of GDP. Second, we use these estimates of GDP to assess the impact of the 2008 financial crisis on the accuracy of agents’ expectations about the evolution of the economic activity in four Scandinavian economies. While we find an improvement in the capacity of agents’ to anticipate economic growth after the crisis, predictive accuracy worsens in relation to the period prior to the crisis. The most accurate GDP forecasts are obtained for Sweden., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

From bilingual to multilingual neural machine translation by incremental training

  • Escolano Peinado, Carlos
  • Ruiz Costa-Jussà, Marta
  • Rodríguez Fonollosa, José Adrián
Multilingual Neural Machine Translation approaches are based on the use of task specific models and the addition of one more language can only be done by retraining the whole system. In this work, we propose a new training schedule that allows the system to scale to more languages without modification of the previous components based on joint training and language-independent encoder/decoder modules allowing for zero-shot translation. This work in progress shows close results to state-of-the-art in the WMT task., This work is supported in part by a Google Faculty Research Award. This work is also supported in part by the Spanish Ministerio de Economa y Competitividad, the European Regional Development Fund and the Agencia Estatal de Investigacin, through the postdoctoral senior grant Ramon y Cajal, contract TEC2015-69266-P (MINECO/FEDER,EU) and contract PCIN-2017- 079 (AEI/MINECO)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Auto-encoding nearest neighbor i-vectors for speaker verification

  • Khan, Umair
  • India Massana, Miquel Àngel
  • Hernando Pericás, Francisco Javier
In the last years, i-vectors followed by cosine or PLDA scoringtechniques were the state-of-the-art approach in speaker veri-fication. PLDA requires labeled background data, and thereexists a significant performance gap between the two scoringtechniques. In this work, we propose to reduce this gap by us-ing an autoencoder to transform i-vector into a new speaker vec-tor representation, which will be referred to as ae-vector. Theautoencoder will be trained to reconstruct neighbor i-vectors in-stead of the same training i-vectors, as usual. These neighbori-vectors will be selected in an unsupervised manner accordingto the highest cosine scores to the training i-vectors. The evalua-tion is performed on the speaker verification trials of VoxCeleb-1 database. The experiments show that our proposed ae-vectorsgain a relative improvement of 42% in terms of EER comparedto the conventional i-vectors using cosine scoring, which fillsthe performance gap between cosine and PLDA scoring tech-niques by 92%, but without using speaker labels, Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

CATOTRON – A Neural text-to-speech system in catalan

  • Külebi, Baybars
  • Öktem, Alp
  • Peiró Lilja, Àlex
  • Pascual, Santiago
  • Farrús, Mireia
Comunicació presentada a Interspeech 2020 celebrat del 25 al 29 d'octubre de 2020 a Shanghai, Xina., We present Catotron, a neural network-based open-source speech synthesis system in Catalan. Catotron consists of a sequence-to-sequence model trained with two small opensource datasets based on semi-spontaneous and read speech. We demonstrate how a neural TTS can be built for languages with limited resources using found-data optimization and crosslingual transfer learning. We make the datasets, initial models and source code publicly available for both commercial and research purposes., This work was subsidised by the Catalan Department of Culture. A part of the funding comes from the financing administered by the inheritance board of the Generalitat de Catalunya. The last author has been funded by the Agencia Estatal de Investigaci ´on (AEI), Ministerio de Ciencia, Innovaci´on y Universidades and the Fondo Social Europeo (FSE) under grant RYC- 2015-17239 (AEI/FSE, UE). Part of this work has been carried out using an NVIDIA GPU Titan Xp generously provided by NVIDIA Company. This research was also partially supported by the project TEC2015-69266-P (MINECO/FEDER, UE).
Recercat. Dipósit de la Recerca de Catalunya
Publication

Joint model-based recognition and localization of overlapped acoustic events using a set of distributed small microphone arrays

  • Chakraborty, Rupayan
  • Nadeu Camprubí, Climent
Article publicat a Arxiv, sense revisió per parells, In the analysis of acoustic scenes, often the occurring sounds have to be detected in time, recognized, and localized in space. Usually, each of these tasks is done separately. In this paper, a model-based approach to jointly carry them out for the case of multiple simultaneous sources is presented and tested. The recognized event classes and their respective room positions are obtained with a single system that maximizes the combination of a large set of scores, each one resulting from a different acoustic event model and a different beamformer output signal, which comes from one of several arbitrarily-located small microphone arrays. By using a two-step method, the experimental work for a specific scenario consisting of meeting-room acoustic events, either isolated or overlapped with speech, is reported. Tests carried out with two datasets show the advantage of the proposed approach with respect to some usual techniques, and that the inclusion of estimated priors brings a further performance improvement., Preprint
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Multilingual and interlingual semantic representations for natural language processing: a brief introduction

  • Ruiz Costa-Jussà, Marta
  • España-Bonet, Cristina
  • Fung, Pascale
  • Smith, Noah
We introduce the Computational Linguistics special issue on Multilingual and Interlingual Semantic Representations for Natural Language Processing. We situate the special issue’s five articles in the context of our fast-changing field, explaining our motivation for this project. We offer a brief summary of the work in the issue, which includes developments on lexical and sentential semantic representations, from symbolic and neural perspectives., MRC is supported in part by a Google Faculty Research Award 2018, Spanish Ministerio de Economía y Competitividad, the European Regional Development Fund and the Agencia Estatal de Investigacion, through the postdoctoral senior grant Ramon y Cajal, the contract TEC2015-69266-P (MINECO/FEDER,EU) and the contract PCIN-2017-079 (AEI/MINECO). N. A. S. is supported by National Science Foundation grant IIS-1562364. C. E. B. is funded by the German Federal Ministry of Education and Research under the funding code 01IW17001 (Deeplee)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Multi-output RNN-LSTM for multiple speaker speech synthesis with a-interpolation model

  • Pascual, Santiago
  • Bonafonte Cávez, Antonio
Deep Learning has been applied successfully to speech processing. In this paper we propose an architecture for speech synthesis using multiple speakers. Some hidden layers are shared by all the speakers, while there is a specific output layer for each speaker. Objective and perceptual experiments prove that this scheme produces much better results in comparison with sin- gle speaker model. Moreover, we also tackle the problem of speaker interpolation by adding a new output layer (a-layer) on top of the multi-output branches. An identifying code is injected into the layer together with acoustic features of many speakers. Experiments show that the a-layer can effectively learn to interpolate the acoustic features between speakers., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Terminology-aware segmentation and domain feature for the WMT19 biomedical translation task

  • Carrino, Casimiro Pio
  • Rafieian, Bardia
  • Ruiz Costa-Jussà, Marta
  • Rodríguez Fonollosa, José Adrián
In this work, we give a description of the TALP-UPC systems submitted for the WMT19 Biomedical Translation Task. Our proposed strategy is NMT model-independent and relies only on one ingredient, a biomedical terminology list. We first extracted such a terminology list by labelling biomedical words in our training dataset using the BabelNet API. Then, we designed a data preparation strategy to insert the terms information at a token level. Finally, we trained the Transformer model with this terms-informed data. Our best-submitted system ranked 2nd and 3rd for Spanish-English and English-Spanish translation directions, respectively., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Evaluating the underlying gender bias in contextualized word embeddings

  • Basta, Christine Raouf Saad
  • Ruiz Costa-Jussà, Marta
  • Casas Manzanares, Noé
Gender bias is highly impacting natural language processing applications. Word embeddings have clearly been proven both to keep and amplify gender biases that are present in current data sources. Recently, contextualized word embeddings have enhanced previous word embedding techniques by computing word vector representations dependent on the sentence they appear in. In this paper, we study the impact of this conceptual change in the word embedding computation in relation with gender bias. Our analysis includes different measures previously applied in the literature to standard word embeddings. Our findings suggest that contextualized word embeddings are less biased than standard ones even when the latter are debiased., We want to thank Hila Gonen for her support dur-ing our research.This work is supported in part by the Catalan Agency for Management of University andResearch Grants (AGAUR) through the FI PhDScholarship and the Industrial PhD Grant. Thiswork is also supported in part by the Span-ish Ministerio de Economa y Competitividad, the European Regional Development Fund andthe Agencia Estatal de Investigacin, through thepostdoctoral senior grant Ramn y Cajal, contract TEC2015-69266-P (MINECO/FEDER,EU)and contract PCIN-2017-079 (AEI/MINECO)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Self multi-head attention for speaker recognition

  • India Massana, Miquel Àngel
  • Safari, Pooyan
  • Hernando Pericás, Francisco Javier
Most state-of-the-art Deep Learning (DL) approaches forspeaker recognition work on a short utterance level. Given thespeech signal, these algorithms extract a sequence of speakerembeddings from short segments and those are averaged to ob-tain an utterance level speaker representation. In this work wepropose the use of an attention mechanism to obtain a discrim-inative speaker embedding given non fixed length speech utter-ances. Our system is based on a Convolutional Neural Network(CNN) that encodes short-term speaker features from the spec-trogram and a self multi-head attention model that maps theserepresentations into a long-term speaker embedding. The atten-tion model that we propose produces multiple alignments fromdifferent subsegments of the CNN encoded states over the se-quence. Hence this mechanism works as a pooling layer whichdecides the most discriminative features over the sequence toobtain an utterance level representation. We have tested thisapproach for the verification task for the VoxCeleb1 dataset.The results show that self multi-head attention outperforms bothtemporal and statistical pooling methods with a18%of rela-tive EER. Obtained results show a58%relative improvementin EER compared to i-vector+PLDA, Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Language modelling for speaker diarization in telephonic interviews

  • India Massana, Miquel Àngel
  • Hernando Pericás, Francisco Javier
  • Rodríguez Fonollosa, José Adrián
The aim of this paper is to investigate the benefit of combining both language and acoustic modelling for speaker diarization. Although conventional systems only use acoustic features, in some scenarios linguistic data contain high discriminative speaker information, even more reliable than the acoustic ones. In this study we analyze how an appropriate fusion of both kind of features is able to obtain good results in these cases. The proposed system is based on an iterative algorithm where a LSTM network is used as a speaker classifier. The network is fed with character-level word embeddings and a GMM based acoustic score created with the output labels from previous iterations. The presented algorithm has been evaluated in a Call-Center database, which is composed of telephone interview audios. The combination of acoustic features and linguistic content shows a 84.29% improvement in terms of a word-level DER as compared to a HMM/VB baseline system. The results of this study confirms that linguistic content can be efficiently used for some speaker recognition tasks., This work was partially supported by the Spanish Project DeepVoice (TEC2015-69266-P) and by the project PID2019-107579RBI00/ AEI /10.13039/501100011033., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Towards large scale multimedia indexing: a case study on person discovery in broadcast news

  • Le, Nam
  • Bredin, Hervé
  • Sargent, Gabriel
  • India, Miquel
  • López Otero, Paula
  • Barras, Claude
  • Guinaudeau, Camille
  • Gravier, Guillaume
  • Barbosa da Fonseca, Gabriel
  • Lyon Freire, Izabela
  • Patrocínio, Zenilton
  • Guimarães, Silvio Jamil F.
  • Martí Juan, Gerard
  • Morros, Josep Ramon
  • Hernando, Javier
  • Docio Fernández, Laura
  • García Mateo, Carmen
  • Meignier, Sylvain
  • Odobez, Jean-Marc
Comunicació presentada a: the 15th International Workshop on Content-Based Multimedia Indexing (CBMI'17), celebrat a Florència, Itàlia, del 19 al 21 de juny de 2017, The rapid growth of multimedia databases and the human interest in their peers make indices representing the location and identity of people in audio-visual documents essential for searching archives. Person discovery in the absence of prior identity knowledge requires accurate association of audio-visual cues and detected names. To this end, we present 3 different strategies to approach this problem: clustering-based naming, verification-based naming, and graph-based naming. Each of these strategies utilizes different recent advances in unsupervised face / speech representation, verification, and optimization. To have a better understanding of the approaches, this paper also provides a quantitative and qualitative comparative study of these approaches using the associated corpus of the Person Discovery challenge at MediaEval 2016. From the results of our experiments, we can observe the pros and cons of each approach, thus paving the way for future promising research directions., This work was supported by the EU project EUMSSI (FP7-611057), ANR project MetaDaTV (ANR-14-CE24-0024) project, Camomile project (PCIN-2013-067), and the projects TEC2013-43935-R, TEC2015-69266-P, TEC2016-75976-R, TEC2015-65345-P financed by the Spanish government and ERDF.
Recercat. Dipósit de la Recerca de Catalunya
Publication

Let the data do the talking: empirical modelling of survey-based expectations by means of genetic programming

  • Claveria González, Oscar
  • Monte Moreno, Enrique
  • Torra Porras, Salvador
Working paper, In this study we use agents’ expectations about the state of the economy to generate indicators of economic activity in twenty-six European countries grouped in five regions (Western, Eastern, and Southern Europe, and Baltic and Scandinavian countries). We apply a data-driven procedure based on evolutionary computation to transform survey variables in economic growth rates. In a first step, we design five independent experiments to derive the optimal combination of expectations that best replicates the evolution of economic growth in each region by means of genetic programming, limiting the integration schemes to the main mathematical operations. We then rank survey variables according to their performance in tracking economic activity, finding that agents’ “perception about the overall economy compared to last year” is the survey variable with the highest predictive power. In a second step, we assess the out-of-sample forecast accuracy of the evolved indicators. Although we obtain different results across regions, Austria, Slovakia, Portugal, Lithuania and Sweden are the economies of each region that show the best forecast results. We also find evidence that the forecasting performance of the survey-based indicators improves during periods of higher growth., Preprint
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

DNN speaker embeddings using autoencoder pre-training

  • Khan, Umair
  • Hernando Pericás, Francisco Javier
Over the last years, i-vectors have been the state-of-the-art approach in speaker recognition. Recent improvements in deep learning have increased the discriminative quality of i-vectors. However, deep learning architectures require a large amount of labeled background data which is difficult in practice. The aim of this paper is to propose an alternative scheme in order to reduce the need of labeled data. We propose the use of autoencoder pre-training in a speaker verification task. First, we train an autoencoder in an unsupervised way, using a large amount of unlabeled background data. Then, we train a Deep Neural Network (DNN) initialized with the parameters of the pre-trained autoencoder. The DNN training is carried out in a supervised way using relatively small labeled background data. In the testing phase, we extract speaker embeddings as the output of an intermediate layer of the DNN. The training and evaluation were performed on VoxCeleb-2 and VoxCeleb1 databases, respectively. The experimental results have shown that by initializing DNN with the parameters of the pre-trained autoencoder, we have achieved a relative improvement of 21%, in terms of Equal Error Rate (EER), over the baseline i-vector/PLDA system., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Unsupervised training of siamese networks for speaker verification

  • Khan, Umair
  • Hernando Pericás, Francisco Javier
Speaker labeled background data is an essential requirement for most state-of-the-art approaches in speaker recognition, e.g., xvectors and i-vector/PLDA. However, in reality it is difficult to access large amount of labeled data. In this work, we propose siamese networks for speaker verification without using speaker labels. We propose two different siamese networks having two and three branches, respectively, where each branch is a CNN encoder. Since the goal is to avoid speaker labels, we propose to generate the training pairs in an unsupervised manner. The client samples are selected within one database according to highest cosine scores with the anchor in i-vector space. The impostor samples are selected in the same way but from another database. Our double-branch siamese performs binary classification using cross entropy loss during training. In testing phase, we obtain speaker verification scores directly from its output layer. Whereas, our triple-branch siamese is trained to learn speaker embeddings using triplet loss. During testing, we extract speaker embeddings from its output layer, which are scored in the experiments using cosine scoring. The evaluation is performed on VoxCeleb-1 database, which show that using the proposed unsupervised systems, solely or in fusion, the results get closer to supervised baseline, This work has been developed in the framework of DeepVoice Project (TEC2015-69266-P), funded by Spanish Ministry, Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Time series features and machine learning forecasts

  • Claveria González, Oscar
  • Monte Moreno, Enrique
  • Torra Porras, Salvador
In this study we combine the results of two independent analyses to position Spanish regions according to both the characteristics of the time series of international tourist arrivals and the accuracy of predictions of arrivals at the regional level. We apply a seasonal-trend decomposition procedure based on non-parametric regression to isolate the different components of the series and calculate the main time series features. Predictions are generated with several machine learning models in a recursive multi-step-ahead forecasting experiment. Finally, we summarize all the information from the two previous experiments using categorical principal component analysis. By overlapping the distribution of the regions and the component loadings of each variable along both dimensions, we observe that entropy and dispersion show an inverse relation with forecast accuracy, but the interactions between the rest of the features and accuracy are heavily dependent on the forecast horizon. On this evidence, we conclude that in order to increase forecast accuracy of tourist arrivals at the regional level, model selection should be region-specific and based on the forecast horizon., This research was supported by by the projects ECO2016-75805-R and TEC2015-69266-P from the Spanish Ministry of Economy and Competitiveness. We would like to thank the Editor and two anonymous referees for their useful comments and suggestions., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Spanish statistical parametric speech synthesis using a neural vocoder

  • Bonafonte Cávez, Antonio
  • Pascual de la Puente, Santiago
  • Dorca, G.
During the 2000s decade, unit-selection based text-to-speech was the dominant commercial technology. Meanwhile, the TTS research community has made a big effort to push statistical-parametric speech synthesis to get similar quality and more flexibility on the synthetically generated voice. During last years, deep learning advances applied to speech synthesis have filled the gap, specially when neural vocoders substitute traditional signal-processing based vocoders. In this paper we propose to substitute the waveform generation vocoder of MUSA, our Spanish TTS, with SampleRNN, a neural vocoder which was recently proposed as a deep autoregressive raw waveform generation model. MUSA uses recurrent neural networks to predict vocoder parameters (MFCC and logF0) from linguistic features. Then, the Ahocoder vocoder is used to recover the speech waveform out of the predicted parameters. In the first system SampleRNN is extended to generate speech conditioned on the Ahocoder generated parameters (mfcc and logF0), where two configurations have been considered to train the system. First, the parameters derived from the signal using Ahocoder are used. Secondly, the system is trained with the parameters predicted by MUSA, where SampleRNN and MUSA are jointly optimized. The subjective evaluation shows that the second system outperforms both the original Ahocoder and SampleRNN as an independent neural vocoder., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Automatic speech recognition with deep neural networks for impaired speech

  • España-i-Bonet, Cristina
  • Rodríguez Fonollosa, José Adrián
The final publication is available at https://link.springer.com/chapter/10.1007%2F978-3-319-49169-1_10, Automatic Speech Recognition has reached almost human performance in some controlled scenarios. However, recognition of impaired speech is a difficult task for two main reasons: data is (i) scarce and (ii) heterogeneous. In this work we train different architectures on a database of dysarthric speech. A comparison between architectures shows that, even with a small database, hybrid DNN-HMM models outperform classical GMM-HMM according to word error rate measures. A DNN is able to improve the recognition word error rate a 13% for subjects with dysarthria with respect to the best classical architecture. This improvement is higher than the one given by other deep neural networks such as CNNs, TDNNs and LSTMs. All the experiments have been done with the Kaldi toolkit for speech recognition for which we have adapted several recipes to deal with dysarthric speech and work on the TORGO database. These recipes are publicly available., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Towards mitigating gender bias in a decoder-based neural machine translation model by adding contextual information

  • Basta, Christine Raouf Saad
  • Ruiz Costa-Jussà, Marta
  • Rodríguez Fonollosa, José Adrián
Gender bias negatively impacts many natural language processing applications, including ma-chine translation (MT). The motivation behind this work is to study whether recent proposedMT techniques are significantly contributing to attenuate biases in document-level and gender-balanced data. For the study, we consider approaches of adding the previous sentence and thespeaker information, implemented in a decoder-based neural MT system. We show improve-ments both in translation quality (+1 BLEU point) as well as in gender bias mitigation onWinoMT (+5% accuracy)., This work is supported in part by the Catalan Agency for Management of University and Research Grants (AGAUR) through the FI PhD Scholarship. This work is also supported in part by the Spanish Ministerio de Economía y Competitividad, the European Regional Development Fund and the Agencia Estatal de Investigacion, through the postdoctoral senior grant Ramón y Cajal, contract TEC2015-69266-P (MINECO/FEDER,EU) and contract PCIN-2017-079 (AEI/MINECO)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Economic uncertainty: a geometric indicator of discrepancy among experts’ expectations

  • Claveria González, Oscar
  • Monte Moreno, Enrique
  • Torra Porras, Salvador
The final publication is available at Springer via http://dx.doi.org/10.1007/s11205-018-1984-2, In this study we present a geometric approach to proxy economic uncertainty. We design a positional indicator of disagreement among survey-based agents’ expectations about the state of the economy. Previous dispersion-based uncertainty indicators derived from business and consumer surveys exclusively make use of the two extreme pieces of information: the percentage of respondents expecting a variable to rise and to fall. With the aim of also incorporating the information coming from the share of respondents expecting a variable to remain constant, we propose a geometrical framework and use a barycentric coordinate system to generate a measure of disagreement, referred to as a discrepancy indicator. We assess its performance both empirically and experimentally by comparing it to the standard deviation of the share of positive and negative responses. When applied in sixteen European countries, we find that both time-varying metrics co-evolve in most countries for expectations about the country’s overall economic situation in the present, but not in the future. Additionally, we obtain their simulated sampling distributions and we find that the proposed indicator gravitates uniformly towards the three vertices of the simplex representing the three answering categories, as opposed to the standard deviation, which tends to overestimate the level of uncertainty as a result of ignoring the no-change responses. Consequently, we find evidence that the information coming from agents expecting a variable to remain constant has an effect on the measurement of disagreement., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Refinement of unsupervised cross-lingual word embeddings

  • Biesialska, Magdalena Marta
  • Ruiz Costa-Jussà, Marta
Cross-lingual word embeddings aim to bridge the gap between high-resource and low-resource languages by allowing to learn multilingual word representations even without using any direct bilingual signal. The lion's share of the methods are projection-based approaches that map pre-trained embeddings into a shared latent space. These methods are mostly based on the orthogonal transformation, which assumes language vector spaces to be isomorphic. However, this criterion does not necessarily hold, especially for morphologically-rich languages. In this paper, we propose a self-supervised method to refine the alignment of unsupervised bilingual word embeddings. The proposed model moves vectors of words and their corresponding translations closer to each other as well as enforces length- and center-invariance, thus allowing to better align cross-lingual embeddings. The experimental results demonstrate the effectiveness of our approach, as in most cases it outperforms state-of-the-art methods in a bilingual lexicon induction task., We thank anonymous reviewers for their helpful comments. This work is supported in part by the Spanish Ministerio de Economía y Competitividad, the European Regional Development Fund and the Agencia Estatal de Investigación, through the post-doctoral senior grant Ramón y Cajal, the contract TEC2015-69266-P (MINECO/FEDER,EU) and the contract PCIN-2017-079 (AEI/MINECO)., Peer Reviewed, Postprint (published version)