Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 14
Encontrada(s) 1 página(s)

Biblos-e Archivo. Repositorio Institucional de la UAM
Publication Artículo científico (JournalArticle)

Evaluating posterior probabilities: decision theory, proper scoring rules, and calibration

  • Ferrer, Luciana
  • Ramos Castro, Daniel
Most machine learning classifiers are designed to output posterior probabilities for the classes given the input sample. These probabilities may be used to make the categorical decision on the class of the sample; provided as input to a downstream system; or provided to a human for interpretation. Evaluating the quality of the posteriors generated by these system is an essential problem which was addressed decades ago with the invention of proper scoring rules (PSRs). Unfortunately, much of the recent machine learning literature uses calibration metrics—most commonly, the expected calibration error (ECE)—as a proxy to assess posterior performance. The problem with this approach is that calibration metrics reflect only one aspect of the quality of the posteriors, ignoring the discrimination performance. For this reason, we argue that calibration metrics should play no role in the assessment of posterior quality. Expected PSRs should instead be used for this job, preferably normalized for ease of interpretation. In this work, we first give a brief review of PSRs from a practical perspective, motivating their definition using Bayes decision theory. We discuss why expected PSRs provide a principled measure of the quality of a system’s posteriors and why calibration metrics are not the right tool for this job. We argue that calibration metrics, while not useful for performance assessment, may be used as diagnostic tools during system development. With this purpose in mind, we discuss a simple and practical calibration metric, called calibration loss, derived from a decomposition of expected PSRs. We compare this metric with the ECE and with the expected score divergence calibration metric from the PSR literature and argue, using theoretical and empirical evidence, that calibration loss is superior to these two metrics, This material is based upon work supported by the European Union’s Horizon 2020 research and innovation program under grant No 101007666/ESPERANTO/H2020-MSCA-RISE-2020. We thank Niko Brümmer for many enlightening discussions about calibration and PSRs, for his valuable feedback on this paper, and for contributing the proof in Section 2.4.
Addi. Archivo Digital para la Docencia y la Investigación
Publication Artículo científico (JournalArticle)

An Overview of the IberSpeech-RTVE 2022 Challenges on Speech Technologies

  • Lleida, Eduardo
  • Rodríguez Fuentes, Luis Javier
  • Tejedor, Javier
  • Ortega, Alfonso
  • Miguel, Antonio
  • Bazán, Virginia
  • Pérez, Carmen
  • de Prada, Alberto
  • Peñagarikano Badiola, Mikel
  • Varona Fernández, Amparo
  • Bordel García, Germán
  • Torre-Toledano, Doroteo
  • Álvarez, Aitor
  • Arzelus, Haritz
Evaluation campaigns provide a common framework with which the progress of speech technologies can be effectively measured. The aim of this paper is to present a detailed overview of the IberSpeech-RTVE 2022 Challenges, which were organized as part of the IberSpeech 2022 conference under the ongoing series of Albayzin evaluation campaigns. In the 2022 edition, four challenges were launched: (1) speech-to-text transcription; (2) speaker diarization and identity assignment; (3) text and speech alignment; and (4) search on speech. Different databases that cover different domains (e.g., broadcast news, conference talks, parliament sessions) were released for those challenges. The submitted systems also cover a wide range of speech processing methods, which include hidden Markov model-based approaches, end-to-end neural network-based methods, hybrid approaches, etc. This paper describes the databases, the tasks and the performance metrics used in the four challenges. It also provides the most relevant features of the submitted systems and briefly presents and discusses the obtained results. Despite employing state-of-the-art technology, the relatively poor performance attained in some of the challenges reveals that there is still room for improvement. This encourages us to carry on with the Albayzin evaluation campaigns in the coming years., This work was partially supported by Radio Televisión Española through the RTVE Chair at the University of Zaragoza, and Red Temática en Tecnologías del Habla (RED2022-134270-T), funded by AEI (Ministerio de Ciencia e Innovación); It was also partially funded by the European Union’s Horizon 2020 research and innovation program under Marie Skłodowska-Curie Grant 101007666; in part by MCIN/AEI/10.13039/501100011033 and by the European Union “NextGenerationEU”/ PRTR under Grants PDC2021-120846C41 PID2021-126061OB-C44, and in part by the Government of Aragon (Grant Group T3623R); it was also partially funded by the Spanish Ministry of Science and Innovation (OPEN-SPEECH project, PID2019-106424RB-I00) and by the Basque Government under the general support program to research groups (IT-1704-22), and by projects RTI2018-098091-B-I00 and PID2021-125943OB-I00 (Spanish Ministry of Science and Innovation and ERDF) as well.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

The Domain Mismatch Problem in the Broadcast Speaker Attribution Task

  • Viñals, Ignacio
  • Ortega, Alfonso
  • Miguel, Antonio
  • Lleida, Eduardo
The demand of high-quality metadata for the available multimedia content requires the development of new techniques able to correctly identify more and more information, including the speaker information. The task known as speaker attribution aims at identifying all or part of the speakers in the audio under analysis. In this work, we carry out a study of the speaker attribution problem in the broadcast domain. Through our experiments, we illustrate the positive impact of diarization on the final performance. Additionally, we show the influence of the variability present in broadcast data, depicting the broadcast domain as a collection of subdomains with particular characteristics. Taking these two factors into account, we also propose alternative approximations robust against domain mismatch. These approximations include a semisupervised alternative as well as a totally unsupervised new hybrid solution fusing diarization and speaker assignment. Thanks to these two approximations, our performance is boosted around a relative 50%. The analysis has been carried out using the corpus for the Albayzín 2020 challenge, a diarization and speaker attribution evaluation working with broadcast data. These data, provided by Radio Televisión Española (RTVE), the Spanish public Radio and TV Corporation, include multiple shows and genres to analyze the impact of new speech technologies in real-world scenarios.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

aDCF loss function for deep metric learning in end-to-end text-dependent speaker verification systems

  • Mingote, V.
  • Miguel, A.
  • Ribas, D.
  • Ortega, A.
  • Lleida, E.
Metric learning approaches have widely expanded to the training of Speaker Verification (SV) systems based on Deep Neural Networks (DNNs), by using a loss function more consistent with the evaluation process than the traditional identification losses. However, these methods do not consider the performance measure and can involve high computational cost, for example, the need for a careful pair or triplet data selection. This paper proposes the approximated Detection Cost Function (aDCF) loss, which is a loss function based on the measure of the decision errors in SV systems, namely the False Rejection Rate (FRR) and the False Acceptance Rate (FAR). With aDCF loss as the training objective function, the end-to-end system learns how to minimize decision errors. Furthermore, we replace the typical linear layer as the last layer of DNN by a cosine distance layer, which reduces the difference between the metric in the training process and the metric during evaluation. aDCF loss function was evaluated in RSR2015-Part I and RSR2015-Part II datasets for text-dependent speaker verification. The system trained with aDCF loss outperforms all the state-of-the-art functions employed in this paper in both parts of the database.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Multimodal Diarization Systems by Training Enrollment Models as Identity Representations

  • Mingote, Victoria
  • Viñals, Ignacio
  • Gimeno, Pablo
  • Miguel, Antonio
  • Ortega, Alfonso
  • Lleida, Eduardo
This paper describes a post-evaluation analysis of the system developed by ViVoLAB research group for the IberSPEECH-RTVE 2020 Multimodal Diarization (MD) Challenge. This challenge focuses on the study of multimodal systems for the diarization of audiovisual files and the assignment of an identity to each segment where a person is detected. In this work, we implemented two different subsystems to address this task using the audio and the video from audiovisual files separately. To develop our subsystems, we used the state-of-the-art speaker and face verification embeddings extracted from publicly available deep neural networks (DNN). Different clustering techniques were also employed in combination with the tracking and identity assignment process. Furthermore, we included a novel back-end approach in the face verification subsystem to train an enrollment model for each identity, which we have previously shown to improve the results compared to the average of the enrollment data. Using this approach, we trained a learnable vector to represent each enrollment character. The loss function employed to train this vector was an approximated version of the detection cost function (aDCF) which is inspired by the DCF widely used metric to measure performance in verification tasks. In this paper, we also focused on exploring and analyzing the effect of training this vector with several configurations of this objective loss function. This analysis allows us to assess the impact of the configuration parameters of the loss in the amount and type of errors produced by the system.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Unsupervised adaptation of deep speech activity detection models to unseen domains

  • Gimeno, P.
  • Ribas, D.
  • Ortega, A.
  • Miguel, A.
  • Lleida, E.
Speech Activity Detection (SAD) aims to accurately classify audio fragments containing human speech. Current state-of-the-art systems for the SAD task are mainly based on deep learning solutions. These applications usually show a significant drop in performance when test data are different from training data due to the domain shift observed. Furthermore, machine learning algorithms require large amounts of labelled data, which may be hard to obtain in real applications. Considering both ideas, in this paper we evaluate three unsupervised domain adaptation techniques applied to the SAD task. A baseline system is trained on a combination of data from different domains and then adapted to a new unseen domain, namely, data from Apollo space missions coming from the Fearless Steps Challenge. Experimental results demonstrate that domain adaptation techniques seeking to minimise the statistical distribution shift provide the most promising results. In particular, Deep CORAL method reports a 13% relative improvement in the original evaluation metric when compared to the unadapted baseline model. Further experiments show that the cascaded application of Deep CORAL and pseudo-labelling techniques can improve even more the results, yielding a significant 24% relative improvement in the evaluation metric when compared to the baseline system.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Wiener Filter and Deep Neural Networks: A Well-Balanced Pair for Speech Enhancement

  • Ribas, Dayana
  • Miguel, Antonio
  • Ortega, Alfonso
  • Lleida, Eduardo
This paper proposes a Deep Learning (DL) based Wiener filter estimator for speech enhancement in the framework of the classical spectral-domain speech estimator algorithm. According to the characteristics of the intermediate steps of the speech enhancement algorithm, i.e., the SNR estimation and the gain function, there is determined the best usage of the network at learning a robust instance of the Wiener filter estimator. Experiments show that the use of data-driven learning of the SNR estimator provides robustness to the statistical-based speech estimator algorithm and achieves performance on the state-of-the-art. Several objective quality metrics show the performance of the speech enhancement and beyond them, there are examples of noisy vs. enhanced speech available for listening to demonstrate in practice the skills of the method in simulated and real audio.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Class token and knowledge distillation for multi-head self-attention speaker verification systems

  • Mingote, Victoria
  • Miguel, Antonio
  • Ortega, Alfonso
  • Lleida, Eduardo
This paper explores three novel approaches to improve the performance of speaker verification (SV) systems based on deep neural networks (DNN) using Multi-head Self-Attention (MSA) mechanisms and memory layers. Firstly, we propose the use of a learnable vector called Class token to replace the average global pooling mechanism to extract the embeddings. Unlike global average pooling, our proposal takes into account the temporal structure of the input what is relevant for the text-dependent SV task. The class token is concatenated to the input before the first MSA layer, and its state at the output is used to predict the classes. To gain additional robustness, we introduce two approaches. First, we have developed a new sampling estimation of the class token. In this approach, the class token is obtained by sampling from a list of several trainable vectors. This strategy introduces uncertainty that helps to generalize better compared to a single initialization as it is shown in the experiments. Second, we have added a distilled representation token for training a teacher-student pair of networks using the Knowledge Distillation (KD) philosophy, which is combined with the class token. This distillation token is trained to mimic the predictions from the teacher network, while the class token replicates the true label. All the strategies have been tested on the RSR2015-Part II and DeepMine-Part 1 databases for text-dependent SV, providing competitive results compared to the same architecture using the average pooling mechanism to extract average embeddings.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Automatic voice disorder detection using self-supervised representations

  • Ribas, Dayana
  • Pastor, Miguel A.
  • Miguel, Antonio
  • Martinez, David
  • Ortega, Alfonso
  • Lleida, Eduardo
Many speech features and models, including Deep Neural Networks (DNN), are used for classification tasks between healthy and pathological speech with the Saarbruecken Voice Database (SVD). However, accuracy values of 80.71% for phrases or 82.8% for vowels /aiu/ are the highest reported for audio samples in SVD when the evaluation includes the wide amount of pathologies in the database, instead of a selection of some pathologies. This paper targets this top performance in the state-of-the-art Automatic Voice Disorder Detection (AVDD) systems. In the framework of a DNN-based AVDD system we study the capability of Self-Supervised (SS) representation learning for describing discriminative cues between healthy and pathological speech. The system processes the SS temporal sequence of features with a single feed-forward layer and Class-Token (CT) Transformer for obtaining the classification between healthy and pathological speech. Furthermore, there is evaluated a suitable data extension of the training set with out-of-domain data is also evaluated to deal with the low availability of data for using DNN-based models in voice pathology detection. Experimental results using audio samples corresponding to phrases in the SVD dataset, including all pathologies available, show classification accuracy values until 93.36%. This means that the proposed AVDD system achieved accuracy improvements of 4.1% without the training data extension, and 15.62% after the training data extension compared to the baseline system. Beyond the novelty of using SS representations for AVDD, the fact of obtaining accuracies over 90% in these conditions and using the whole set of pathologies in the SVD is a milestone for voice disorder-related research. Furthermore, the study on the amount of in-domain data in the training set related to the system performance show guidance for the data preparation stage. Lessons learned in this work suggest guidelines for taking advantage of DNN, to boost the performance in developing automatic systems for diagnosis, treatment, and monitoring of voice pathologies.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Cross-corpus training strategy for speech emotion recognition using self-supervised representations

  • Pastor, Miguel A.
  • Ribas, Dayana
  • Ortega, Alfonso
  • Miguel, Antonio
  • Lleida, Eduardo
Speech Emotion Recognition (SER) plays a crucial role in applications involving human-machine interaction. However, the scarcity of suitable emotional speech datasets presents a major challenge for accurate SER systems. Deep Neural Network (DNN)-based solutions currently in use require substantial labelled data for successful training. Previous studies have proposed strategies to expand the training set in this framework by leveraging available emotion speech corpora. This paper assesses the impact of a cross-corpus training extension for a SER system using self-supervised (SS) representations, namely HuBERT and WavLM. The feasibility of training systems with just a few minutes of in-domain audio is also analyzed. The experimental results demonstrate that augmenting the training set with EmoDB (German), RAVDESS, and CREMA-D (English) datasets leads to improved SER accuracy on the IEMOCAP dataset. By combining a cross-corpus training extension and SS representations, state-of-the-art performance is achieved. These findings suggest that the cross-corpus strategy effectively addresses the scarcity of labelled data and enhances the performance of SER systems.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Automatic voice disorder detection from a practical perspective

  • Vidal, Jazmin
  • Ribas, Dayana
  • Bonomi, Cyntia
  • Lleida, Eduardo
  • Ferrer, Luciana
  • Ortega, Alfonso
Voice disorders, such as dysphonia, are common among the general population. These pathologies often remain untreated until they reach a high level of severity. Assisting the detection of voice disorders could facilitate early diagnosis and subsequent treatment. In this study, we address the practical aspects of automatic voice disorders detection (AVDD). In real-world scenarios, data annotated for voice disorders is usually scarce due to various challenges involved in the collection and annotation of such data. However, some relatively large datasets are available for a reduced number of domains. In this context, we propose the use of a combination of out-of-domain and in-domain data for training a deep neural network-based AVDD system, and offer guidance on the minimum amount of in-domain data required to achieve acceptable performance. Further, we propose the use of a cost-based metric, the normalized expected cost (EC), to evaluate performance of AVDD systems in a way that closely reflects the needs of the application. As an added benefit, optimal decisions for the EC can be made in a principled way given by Bayes decision theory. Finally, we argue that for medical applications like AVDD, the categorical decisions need to be accompanied by interpretable scores that reflect the confidence of the system. Even very accurate models often produce scores that are not suited for interpretation. Here, we show that such models can be easily improved by adding a calibration stage-trained with just a few minutes of in-domain data. The outputs of the resulting calibrated system can then better support practitioners in their decision-making process.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Generalizing AUC Optimization to Multiclass Classification for Audio Segmentation With Limited Training Data

  • Gimeno, P
  • Mingote, V
  • Ortega, A
  • Miguel, A
  • Lleida, E
Area under the ROC curve (AUC) optimisation techniques developed for neural networks have recently demonstrated their capabilities in different audio and speech related tasks. However, due to its intrinsic nature, AUC optimisation has focused only on binary tasks so far. In this paper, we introduce an extension to the AUC optimisation framework so that it can be easily applied to an arbitrary number of classes, aiming to overcome the issues derived from training data limitations in deep learning solutions. Building upon the multiclass definitions of the AUC metric found in the literature, we define two new training objectives using a one-versus-one and a one-versus-rest approach. In order to demonstrate its potential, we apply them in an audio segmentation task with limited training data that aims to differentiate 3 classes: foreground music, background music and no music. Experimental results show that our proposal can improve the performance of audio segmentation systems significantly compared to traditional training criteria such as cross entropy.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Shouted and whispered speech compensation for speaker verification systems

  • Prieto, S.
  • Ortega, A.
  • López-Espejo, I.
  • Lleida, E.
Nowadays, speaker verification systems begin to perform very well under normal speech conditions due to the plethora of neutrally-phonated speech data available, which are used to train such systems. Nevertheless, the use of vocal effort modes other than normal severely degrades performance because of vocal effort mismatch. In this paper, in which we consider whispered, normal and shouted speech production modes, we first study how vocal effort mismatch negatively affects speaker verification performance. Then, in order to mitigate this issue, we describe a series of techniques for score calibration and speaker embedding compensation relying on logistic regression-based vocal effort mode detection. To test the validity of all of these methodologies, speaker verification experiments using a modern x-vector-based speaker verification system are carried out. Experimental results show that we can achieve, when combining score calibration and embedding compensation relying upon vocal effort mode detection, up to 19% and 52% equal error rate (EER) relative improvements under the shouted-normal and whispered-normal scenarios, respectively, in comparison with a system applying neither calibration nor compensation. Compared to our previous work 1], we obtain a 7.3% relative improvement in terms of EER when adding score calibration in shouted-normal All vs. All condition. © 2022 Elsevier Inc.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Beyond Global Metrics: A Fairness Analysis for Interpretable Voice Disorder Detection Systems

  • Estevez, Mariel
  • Bonomi, Cyntia
  • Ribas, Dayana
  • Ortega, Alfonso
  • Ferrer, Luciana
We investigate and quantify demographic-dependent biases in automatic voice disorders detection (AVDD) systems by analyzing performance disparities across speaker groups, and evaluate group-specific calibration strategies for improving reliability. We conducted a comprehensive analysis of an AVDD system using existing voice disorder datasets with available demographic metadata. The study involved analyzing system performance across various demographic groups, particularly focusing on gender and age-based cohorts. Performance evaluation was based on multiple metrics, including normalized costs and cross-entropy. We employed calibration techniques trained separately on predefined demographic groups to address group-dependent miscalibration. Analysis revealed significant performance disparities across demographic groups despite strong global metrics. The system showed systematic biases, misclassifying healthy speakers over 55 as having a voice disorder and speakers with disorders aged 14–30 as healthy. Group-specific calibration improved posterior probability quality, reducing overconfidence. For young disordered speakers, low severity scores were identified as contributing to poor system performance. For older speakers, age-related voice characteristics and potential limitations in the pretrained Hubert model used as a feature extractor likely affected results. The study demonstrates that global performance metrics are insufficient for evaluating AVDD system performance. Group-specific analysis may unmask problems in system performance fairly which are hidden within global metrics. Further, group-dependent calibration strategies help mitigate biases, resulting in a more reliable indication of system confidence. These findings emphasize the need for demographic-specific evaluation and calibration in voice disorder detection systems while providing a methodological framework applicable to broader biomedical classification tasks where demographic metadata is available.