Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 17
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Using word graphs as intermediate representation of uttered sentences

  • Gomez, J.A.|||0000-0002-4174-3762
  • Sanchís Arnal, Emilio|||0000-0002-6737-4723
We present an algorithm for building graphs of words as an intermediate representation of uttered sentences. No language model is used. The input data for the algorithm are the pronunciation lexicon organized as a tree and the sequence of acoustic frames. The transition between consecutive units are considered as additional units. Nodes represent discrete instants of time, arcs are labelled with words, and a confidence measure is assigned to each detected word, which is computed by using the phonetic probabilities of the subsequence of acoustic frames used for completing the word. We evaluated the obtained word graphs by searching the path that best matches with the correct sentence and then measuring the word accuracy, i.e. the oracle word accuracy. © 2012 Springer-Verlag., This work was supported by the Spanish MICINN under contract TIN2011-28169-C05-01 and the Vic. d’Investigació of the UPV under contract 20110897.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

ELIRF at MEDIAEVAL 2013: Spoken Web Search Task

  • Gomez, J.A.|||0000-0002-4174-3762
  • Hurtado Oliver, Lluis Felip|||0000-0002-1877-0455
  • Sanchís Arnal, Emilio|||0000-0002-6737-4723
  • Calvo Lance, Marcos
In this paper, we present the systems that the Natural Language Engineering and Pattern Recognition group (ELiRF) has submitted to the MediaEval 2013 Spoken Web Search task. All of them are based on a Subsequence Dynamic Time Warping algorithm and are zero-resources systems., Work funded by the Spanish Government and the E.U. under contract TIN2011-28169-C05 and FPU Grant AP2010- 4193.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Voice-QA: evaluating the impact of misrecognized words on passage retrieval

  • Calvo Lance, Marcos
  • Buscaldi, Davide
  • Rosso, Paolo|||0000-0002-8922-1242
Question Answering is an Information Retrieval task where the query is posed using natural language and the expected result is a concise answer. Voice-activated Question Answering systems represent an interesting application, where the question is formulated by speech. In these systems, an Automatic Speech Recognition module can be used to transcribe the question. Thus, recognition errors may be introduced, producing a significant effect on the answer retrieval process. In this work we study the relationship between some features of misrecognized words and the retrieval results. The features considered are the redundancy of a word in the result set and its inverse document frequency calculated over the collection. The results show that the redundancy of a word may be an important clue on whether an error on it would deteriorate the retrieval results, at least if a closed model is used for speech recognition., This work was carried out in the framework of TextEnterprise (TIN2009-13391-C04-03), Timpano (TIN2011-28169-C05-01), WIQEI IRSES (grant no. 269180) within the FP 7 Marie Curie People, FPU Grant AP2010-4193 from the Spanish Ministerio de Educaci´on (first author), and the Microcluster VLC/Campus on Multimodal Intelligent Systems (third author)
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Exploiting multiple ASR outputs for a spoken language understanding task

  • Calvo Lance, Marcos
  • García-Granada, Fernando|||0000-0003-2213-4213
  • Hurtado Oliver, Lluis Felip|||0000-0002-1877-0455
  • Jiménez-Serrano, Santiago|||0000-0003-2917-6053
  • Sanchís Arnal, Emilio|||0000-0002-6737-4723
In this paper, we present an approach to Spoken Language Understanding, where the input to the semantic decoding process is a composition of multiple hypotheses provided by the Automatic Speech Recognition module. This way, the semantic constraints can be applied not only to a unique hypothesis, but also to other hypotheses that could represent a better recognition of the utterance. To do this, we have developed an algorithm to combine multiple sentences into a weighted graph of words, which is the input to the semantic decoding process. It has also been necessary to develop a specific algorithm to process these graphs of words according to the statistical models that represent the semantics of the task. This approach has been evaluated in a SLU task in Spanish. Results, considering different configurations of ASR outputs, show the better behavior of the system when a combination of hypotheses is considered., This work is partially supported by the Spanish MICINN under contract TIN2011-28169-C05-01, and under FPU Grant AP2010-4193
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

Combining multiple translation systems for spoken language understanding portability

  • García-Granada, Fernando|||0000-0003-2213-4213
  • Hurtado Oliver, Lluis Felip|||0000-0002-1877-0455
  • Sanchís Arnal, Emilio|||0000-0002-6737-4723
  • Segarra Soriano, Encarnación
  • Riccardi, Giussepe
[EN] We are interested in the problem of learning Spoken Language Understanding (SLU) models for multiple target languages. Learning such models requires annotated corpora, and porting to different languages would require corpora with parallel text translation and semantic annotations. In this paper we investigate how to learn a SLU model in a target language starting from no target text and no semantic annotation. Our proposed algorithm is based on the idea of exploiting the diversity (with regard to performance and coverage) of multiple translation systems to transfer statistically stable word-to-concept mappings in the case of the romance language pair, French and Spanish. Each translation system performs differently at the lexical level (wrt BLEU). The best translation system performances for the semantic task are gained from their combination at different stages of the portability methodology. We have evaluated the portability algorithms on the French MEDIA corpus, using French as the source language and Spanish as the target language. The experiments show the effectiveness of the proposed methods with respect to the source language SLU baseline., This work is partially supported by the Spanish MICINN under contract TIN2011-28169-C05-01, and by the Vic. d'Investigacio of the UPV under contracts PAID-00-09 and PAID-06-10 The author work was partially funded by FP7 PORTDIAL project n.296170
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A prosody-based vector-space model of dialog activity for information retrieval

  • Ward,Nigel G.
  • Werner, Steven D.
  • García-Granada, Fernando
  • Sanchís Arnal, Emilio|||0000-0002-6737-4723
Search in audio archives is a challenging problem. Using prosodic information to help find relevant content has been proposed as a complement to word-based retrieval, but its utility has been an open question. We propose a new way to use prosodic information in search, based on a vector-space model, where each point in time maps to a point in a vector space whose dimensions are derived from numerous prosodic features of the local context. Point pairs that are close in this vector space are frequently similar, not only in terms of the dialog activities, but also in topic. Using proximity in this space as an indicator of similarity, we built support for a query-by-example function. Searchers were happy to use this function, and it provided value on a large testset. Prosody-based retrieval did not perform as well as word-based retrieval, but the two sources of information were often non-redundant and in combination they sometimes performed better than either separately., We thank Martha Larson, Alejandro Vega, Steve Renals, Khiet Truong, Olac Fuentes, David Novick, Shreyas Karkhedkar, Luis F. Ramirez, Elizabeth E. Shriberg, Catharine Oertel, Louis-Philippe Morency, Tatsuya Kawahara, Mary Harper, and the anonymous reviewers. This work was supported in part by the National Science Foundation under Grants IIS-0914868 and IIS-1241434 and by the Spanish MEC under contract TIN2011-28169-C05-01.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

A Multi-domain Dialog System to integrate heterogeneous Spoken Dialog Systems

  • Planells Lerma, Joaquín
  • Segarra Soriano, Encarnación
  • Hurtado Oliver, Lluis Felip|||0000-0002-1877-0455
  • Sanchís Arnal, Emilio|||0000-0002-6737-4723
[EN] In this paper, we present an architecture to create a multidomain spoken dialog system with minimum effort by composing heterogeneous pre-existent spoken dialog systems into a new system able to perform richer interactions. A Task Manager acts as a proxy for the different sub-domains and activates one of the systems each turn. The different sub-systems are not aware that they are used in a multi-domain scenario and believe that they are speaking directly to the user. This allows us to add new domains leaving most of the underlying models unmodified and reducing the amount of time and money needed to deploy the application. The proposed architecture has been applied to create a multi-domain system combining three heterogeneous spoken dialog systems (sport facilities booking, weather service and personal calendar) in Spanish. The evaluation with naive real users shows that this is an appropriate approach to develop multi-domain spoken dialog systems., This work is partially supported by the Spanish MICINN under contract TIN2011-28169-C05-01.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

Evaluating Spoken Dialogue Models under the Interactive Pattern Recognition Framework

  • Ghigi, Fabrizio
  • Torres Barañano, Ines
  • Justo, Raquel
  • Benedí Ruiz, José Miguel
[EN] The new Interactive Pattern Recognition (IPR) framework has been proposed to deal with human-machine interaction. In this context a new formulation has been recently defined to represent a Spoken Dialogue System as an IPR problem. In this work this formulation is applied to define graphical models that deal with Spoken Dialogue Systems. The definition of both a Dialogue Manager and a User Model are shown and the estimation of the parameters and smoothing techniques are presented in the paper. These models were evaluated in a dialogue generation task on two very different corpora: Dihana corpus consisting of Spanish spoken dialogues acquired with the Wizard of Oz technique and Let’s Go corpus consisting of spoken dialogues in English between real users and the Ravenclaw dialogue manager developed by CMU. The results obtained show that original and simulated dialogues exhibited very similar behaviours, thus demonstrating the learning capacity of the proposed models in both a controlled Wizard of Oz task and a spoken dialogue system that interacts with real users. This formulation can then be considered as a promising framework to deal with Spoken Dialogue Systems., Work supported by Spanish Ministry of Science under TIN2011-28169-C05-04 and by Basque Government under grant IT685-13.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

A phonetic-based approach to query-by-example spoken term detection

  • Hurtado Oliver, Lluis Felip|||0000-0002-1877-0455
  • Gomez, J.A.|||0000-0002-4174-3762
  • García-Granada, Fernando|||0000-0003-2213-4213
  • Sanchís Arnal, Emilio|||0000-0002-6737-4723
  • Calvo Lance, Marcos
Query-by-Example Spoken Term Detection (QbE-STD) tasks are usually addressed by representing speech signals as a sequence of feature vectors by means of a parametrization step, and then using a pattern matching technique to find the candidate detections. In this paper, we propose a phoneme-based approach in which the acoustic frames are first converted into vectors representing the a posteriori probabilities for every phoneme. This strategy is specially useful when the language of the task is a priori known. Then, we show how this representation can be used for QbE-STD using both a Segmental Dynamic Time Warping algorithm and a graph-based method. The proposed approach has been evaluated with a QbE-STD task in Spanish, and the results show that it can be an adequate strategy for tackling this kind of problems, Work partially supported by the Spanish Ministerio de Economía y Competitividad under contract TIN2011-28169-C05-01 and FPU Grant AP2010-4193, and by the Vic. d’Investigació of the UPV (PAID-06-10)
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Sentiment Analysis in Twitter for Spanish

  • Pla Santamaría, Ferran
  • Hurtado Oliver, Lluis Felip|||0000-0002-1877-0455
This paper describes a SVM-approach for Sentiment Analysis (SA) in Twitter for Spanish. This task was part of the TASS2013 workshop, which is a framework for SA that is focused on the Spanish language. We describe the approach used, and we present an experimental comparison of the approaches presented by the di erent teams that took part in the competition. We also describe the improvements that were added to our system after our participation in the competition. With these improvements, we obtained an accuracy of 62.88% and 70.25% on the SA test set for 5-level and 3-level tasks respectively. To our knowledge, these results are the best results published until now for the SA tasks of the TASS2013 workshop., This work has been funded by the projects, DIANA (MEC TIN2012-38603-C02-01) and Tímpano (MEC TIN2011-28169-C05-01).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

ELIRF at MEDIAEVAL 2013: Similar Segments of Social Speech Task

  • García-Granada, Fernando|||0000-0003-2213-4213
  • Sanchís Arnal, Emilio|||0000-0002-6737-4723
  • Hurtado Oliver, Lluis Felip|||0000-0002-1877-0455
  • Calvo Lance, Marcos
  • Pla Santamaría, Ferran
This paper describes the Natural Language Engineering and Pattern Recognition group (ELiRF) approaches and results towards the Similar Segments of Social Speech Task of Me- diaEval 2013. The task involves finding segments similar to a query segment in a multimedia collection of informal, un- structured dialogs among members of a small community. Our approach has two phases. In a first phase a preprocess of the sentences is performed based on the morphology and semantics of the words. In a second phase, a searching pro- cess based on different distance measures is carried out. This has been done taking the correctly transcribed sentences and the output of an Automatic Speech Recognizer., Work funded by the Spanish Government and the E.U. under the contracts TIN2011-28169-C05 and TIN2012-38603- C02, and FPU Grant AP2010-4193
Investigo. Repositorio Institucional de la Universidade de Vigo
Publicación Artículo científico (JournalArticle)

Albayzín-2014 evaluation: audio segmentation and classification in broadcast news domains

  • Castán, Diego
  • Tavarez, David
  • López Otero, Paula
  • Franco Pedroso, Javier
  • Delgado, Héctor
  • Navas, Eva
  • Docío Fernández, Laura
  • Ramos Castro, Daniel
  • Serrano, Javier
  • Ortega Giménez, Alfonso
  • Lleida Solano, Eduardo
Audio segmentation is important as a pre-processing task to improve the performance of many speech technology tasks and, therefore, it has an undoubted research interest. This paper describes the database, the metric, the systems and the results for the Albayzín-2014 audio segmentation campaign. In contrast to previous evaluations where the task was the segmentation of non-overlapping classes, Albayzín-2014 evaluation proposes the delimitation of the presence of speech, music and/or noise that can be found simultaneously. The database used in the evaluation was created by fusing different media and noises in order to increase the difficulty of the task. Seven segmentation systems from four different research groups were evaluated and combined. Their experimental results were analyzed and compared with the aim of providing a benchmark and showing up the promising directions in this field., Ministerio de Ciencia e Innovación | Ref. TIN2011-28169-C05-02, Ministerio de Economía y Competitividad | Ref. TEC2012-38939-C03
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

On the Use of Deep Feedforward Neural Networks for Automatic Language Identification

  • Lopez-Moreno, Ignacio
  • Gonzalez-Dominguez, Javier
  • Martínez González, David
  • Plchot, Oldrich
  • Gonzalez-Rodriguez, Joaquin
  • Moreno, Pedro
In this work, we present a comprehensive study on the use of deep neural networks (DNNs) for automatic language identification (LID). Motivated by the recent success of using DNNs in acoustic modeling for speech recognition, we adapt DNNs to the problem of identifying the language in a given utterance from its short-term acoustic features. We propose two different DNN- based approaches. In the first one, the DNN acts as an end-to-end LID classifier, receiving as input the speech features and providing as output the estimated probabilities of the target languages. In the second approach, the DNN is used to extract bottleneck features that are then used as inputs for a state-of-the-art i-vector system. Experiments are conducted in two different scenarios: the complete NIST Language Recognition Evaluation dataset 2009 (LRE’09) and a subset of the Voice of America (VOA) data from LRE’09, in which all languages have the same amount of training data. Results for both datasets demonstrate that the DNN-based systems significantly outperform a state-of-art i-vector system when dealing with short-duration utterances. Furthermore, the combination of the DNN-based and the classical i-vector system leads to additional performance improvements (up to 45% of relative improvement in both EER and Cavg on 3s and 10s conditions, respectively).
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Albayzín-2014 evaluation: audio segmentation and classification in broadcast news domains

  • Castán, D.
  • Tavarez, D.
  • Lopez-Otero, P.
  • Franco-Pedroso, J.
  • Delgado, H.
  • Navas, E.
  • Docio-Fernández, L.
  • Ramos, D.
  • Serrano, J.
  • Ortega, A.
  • Lleida, E.
Audio segmentation is important as a pre-processing task to improve the performance of many speech technology tasks and, therefore, it has an undoubted research interest. This paper describes the database, the metric, the systems and the results for the Albayzín-2014 audio segmentation campaign. In contrast to previous evaluations where the task was the segmentation of non-overlapping classes, Albayzín-2014 evaluation proposes the delimitation of the presence of speech, music and/or noise that can be found simultaneously. The database used in the evaluation was created by fusing different media and noises in order to increase the difficulty of the task. Seven segmentation systems from four different research groups were evaluated and combined. Their experimental results were analyzed and compared with the aim of providing a benchmark and showing up the promising directions in this field.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Audio segmentation-by-classification approach based on factor analysis in broadcast news domain

  • Castán, D.
  • Ortega, A.
  • Miguel, A.
  • Lleida, E.
This paper studies a novel audio segmentation-by-classification approach based on factor analysis. The proposed technique compensates the within-class variability by using class-dependent factor loading matrices and obtains the scores by computing the log-likelihood ratio for the class model to a non-class model over fixed-length windows. Afterwards, these scores are smoothed to yield longer contiguous segments of the same class by means of different back-end systems. Unlike previous solutions, our proposal does not make use of specific acoustic features and does not need a hierarchical structure. The proposed method is applied to segment and classify audios coming from TV shows into five different acoustic classes: speech, music, speech with music, speech with noise, and others. The technique is compared to a hierarchical system with specific acoustic features achieving a significant error reduction.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Prosodic features and formant modeling for an ivector-based language recognition system

  • Martinez, D.
  • Lleida, E.
  • Ortega, A.
  • Miguel, A.
The prosody of a language is encoded in syllable length, loudness and pitch. These attributes make humans perceive rhythm, stress and intonation in speech. Depending on the language, these speech properties vary, making language classification possible. On the other hand, formants are the resonance frequencies of the vocal tract, depend heavily on the position adopted by the articulatory organs, and are especially useful to disambiguate vowel sounds. In this paper prosodic and formant information are combined to build a generative language identification system based on Gaussian models fed with iVectors. The system is evaluated on the NIST LRE09 database and the inclusion of formant information gives about 50% relative improvement for the 30 s task over a prosodic system without it. The fusion with a state-of-the-art acoustic system based on shifted delta cepstral coefficients (SDC) shows the complementarity of both approaches.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Segmentation-by-classification system based on factor analysis

  • Castan, D.
  • Ortega, A.
  • Villalba, J.
  • Miguel, A.
  • Lleida, E.
This paper proposes a novel audio segmentation-by-classification system based on Factor Analysis (FA) with a channel compensation matrix for each class and scoring the fixed-length segments as the log-likelihood ratio between class/no-class. The scores are smoothed and the most probable sequence is computed with a Viterbi algorithm. The system described here is designed to segment and classify the audio files coming from broadcast programs into five different classes: speech (SP), speech with noise (SN), speech with music (SM), music (MU) or others (OT). This task was proposed in the Albayzin 2010 evaluation campaign. The system is compared with the winning system of the evaluation achieving lower error rate in SP and SN. These classes represent 3/4 of the total amount of the data. Therefore, the FA segmentation system gets a reduction in the average segmentation error rate.