Juan Rafael Orozco-Arroyave

Abner Hernandez, Paula Andrea Pérez-Toro, Elmar Nöth et al.

State-of-the-art automatic speech recognition (ASR) systems perform well on healthy speech. However, the performance on impaired speech still remains an issue. The current study explores the usefulness of using Wav2Vec self-supervised speech representations as features for training an ASR system for dysarthric speech. Dysarthric speech recognition is particularly difficult as several aspects of speech such as articulation, prosody and phonation can be impaired. Specifically, we train an acoustic model with features extracted from Wav2Vec, Hubert, and the cross-lingual XLSR model. Results suggest that speech representations pretrained on large unlabelled data can improve word error rate (WER) performance. In particular, features from the multilingual model led to lower WERs than filterbanks (Fbank) or models trained on a single language. Improvements were observed in English speakers with cerebral palsy caused dysarthria (UASpeech corpus), Spanish speakers with Parkinsonian dysarthria (PC-GITA corpus) and Italian speakers with paralysis-based dysarthria (EasyCall corpus). Compared to using Fbank features, XLSR-based features reduced WERs by 6.8%, 22.0%, and 7.0% for the UASpeech, PC-GITA, and EasyCall corpus, respectively.

Soroosh Tayebi Arasteh, Mahshad Lotfinia, Paula Andrea Perez-Toro et al.

Speech pathology has impacts on communication abilities and quality of life. While deep learning-based models have shown potential in diagnosing these disorders, the use of sensitive data raises critical privacy concerns. Although differential privacy (DP) has been explored in the medical imaging domain, its application in pathological speech analysis remains largely unexplored despite the equally critical privacy concerns. To the best of our knowledge, this study is the first to investigate DP's impact on pathological speech data, focusing on the trade-offs between privacy, diagnostic accuracy, and fairness. Using a large, real-world dataset of 200 hours of recordings from 2,839 German-speaking participants, we observed a maximum accuracy reduction of 3.85% when training with DP with high privacy levels. To highlight real-world privacy risks, we demonstrated the vulnerability of non-private models to gradient inversion attacks, reconstructing identifiable speech samples and showcasing DP's effectiveness in mitigating these risks. To explore the potential generalizability across languages and disorders, we validated our approach on a dataset of Spanish-speaking Parkinson's disease patients, leveraging pretrained models from healthy English-speaking datasets, and demonstrated that careful pretraining on large-scale task-specific datasets can maintain favorable accuracy under DP constraints. A comprehensive fairness analysis revealed minimal gender bias at reasonable privacy levels but underscored the need for addressing age-related disparities. Our results establish that DP can balance privacy and utility in speech disorder detection, while highlighting unique challenges in privacy-fairness trade-offs for speech data. This provides a foundation for refining DP methodologies and improving fairness across diverse patient groups in real-world deployments.

Abner Hernandez, Paula Andrea Pérez-Toro, Juan Camilo Vásquez-Correa et al.

Collecting speech data is an important step in training speech recognition systems and other speech-based machine learning models. However, the issue of privacy protection is an increasing concern that must be addressed. The current study investigates the use of voice conversion as a method for anonymizing voices. In particular, we train several voice conversion models using self-supervised speech representations including Wav2Vec2.0, Hubert and UniSpeech. Converted voices retain a low word error rate within 1% of the original voice. Equal error rate increases from 1.52% to 46.24% on the LibriSpeech test set and from 3.75% to 45.84% on speakers from the VCTK corpus which signifies degraded performance on speaker verification. Lastly, we conduct experiments on dysarthric speech data to show that speech features relevant to articulation, prosody, phonation and phonology can be extracted from anonymized voices for discriminating between healthy and pathological speech.

Abner Hernandez, Eunjung Yeo, Kwanghee Choi et al.

The limited availability of dysarthric speech data makes cross-lingual detection an important but challenging problem. A key difficulty is that speech representations often encode language-dependent structure that can confound dysarthria detection. We propose a representation-level language shift (LS) that aligns source-language self-supervised speech representations with the target-language distribution using centroid-based vector adaptation estimated from healthy-control speech. We evaluate the approach on oral DDK recordings from Parkinson's disease speech datasets in Czech, German, and Spanish under both cross-lingual and multilingual settings. LS substantially improves sensitivity and F1 in cross-lingual settings, while yielding smaller but consistent gains in multilingual settings. Representation analysis further shows that LS reduces language identity in the embedding space, supporting the interpretation that LS removes language-dependent structure.

Moreno La Quatra, Juan Rafael Orozco-Arroyave, Marco Sabato Siniscalchi · gatech

This work aims to tackle the Parkinson's disease (PD) detection problem from the speech signal in a bilingual setting by proposing an ad-hoc dual-head deep neural architecture for type-based binary classification. One head is specialized for diadochokinetic patterns. The other head looks for natural speech patterns present in continuous spoken utterances. Only one of the two heads is operative accordingly to the nature of the input. Speech representations are extracted from self-supervised learning (SSL) models and wavelet transforms. Adaptive layers, convolutional bottlenecks, and contrastive learning are exploited to reduce variations across languages. Our solution is assessed against two distinct datasets, EWA-DB, and PC-GITA, which cover Slovak and Spanish languages, respectively. Results indicate that conventional models trained on a single language dataset struggle with cross-linguistic generalization, and naive combinations of datasets are suboptimal. In contrast, our model improves generalization on both languages, simultaneously.

Chin-Jou Li, Eunjung Yeo, Kwanghee Choi et al. · cmu

Automatic speech recognition (ASR) for dysarthric speech remains challenging due to data scarcity, particularly in non-English languages. To address this, we fine-tune a voice conversion model on English dysarthric speech (UASpeech) to encode both speaker characteristics and prosodic distortions, then apply it to convert healthy non-English speech (FLEURS) into non-English dysarthric-like speech. The generated data is then used to fine-tune a multilingual ASR model, Massively Multilingual Speech (MMS), for improved dysarthric speech recognition. Evaluation on PC-GITA (Spanish), EasyCall (Italian), and SSNCE (Tamil) demonstrates that VC with both speaker and prosody conversion significantly outperforms the off-the-shelf MMS performance and conventional augmentation techniques such as speed and tempo perturbation. Objective and subjective analyses of the generated data further confirm that the generated speech simulates dysarthric characteristics.

Paula Andrea Pérez-Toro, Tomás Arias-Vergara, Fangxu Xing et al.

Understanding the relationship between vocal tract motion during speech and the resulting acoustic signal is crucial for aided clinical assessment and developing personalized treatment and rehabilitation strategies. Toward this goal, we introduce an audio-to-video generation framework for creating Real Time/cine-Magnetic Resonance Imaging (RT-/cine-MRI) visuals of the vocal tract from speech signals. Our framework first preprocesses RT-/cine-MRI sequences and speech samples to achieve temporal alignment, ensuring synchronization between visual and audio data. We then employ a modified stable diffusion model, integrating structural and temporal blocks, to effectively capture movement characteristics and temporal dynamics in the synchronized data. This process enables the generation of MRI sequences from new speech inputs, improving the conversion of audio into visual data. We evaluated our framework on healthy controls and tongue cancer patients by analyzing and comparing the vocal tract movements in synthesized videos. Our framework demonstrated adaptability to new speech inputs and effective generalization. In addition, positive human evaluations confirmed its effectiveness, with realistic and accurate visualizations, suggesting its potential for outpatient therapy and personalized simulation of vocal tract visualizations.

Soroosh Tayebi Arasteh, Cristian David Rios-Urrego, Elmar Noeth et al.

Parkinson's disease (PD) is a neurological disorder impacting a person's speech. Among automatic PD assessment methods, deep learning models have gained particular interest. Recently, the community has explored cross-pathology and cross-language models which can improve diagnostic accuracy even further. However, strict patient data privacy regulations largely prevent institutions from sharing patient speech data with each other. In this paper, we employ federated learning (FL) for PD detection using speech signals from 3 real-world language corpora of German, Spanish, and Czech, each from a separate institution. Our results indicate that the FL model outperforms all the local models in terms of diagnostic accuracy, while not performing very differently from the model based on centrally combined training sets, with the advantage of not requiring any data sharing among collaborators. This will simplify inter-institutional collaborations, resulting in enhancement of patient outcomes.

Philipp Klumpp, Tomás Arias-Vergara, Paula Andrea Pérez-Toro et al.

Current state of the art acoustic models can easily comprise more than 100 million parameters. This growing complexity demands larger training datasets to maintain a decent generalization of the final decision function. An ideal dataset is not necessarily large in size, but large with respect to the amount of unique speakers, utilized hardware and varying recording conditions. This enables a machine learning model to explore as much of the domain-specific input space as possible during parameter estimation. This work introduces Common Phone, a gender-balanced, multilingual corpus recorded from more than 11.000 contributors via Mozilla's Common Voice project. It comprises around 116 hours of speech enriched with automatically generated phonetic segmentation. A Wav2Vec 2.0 acoustic model was trained with the Common Phone to perform phonetic symbol recognition and validate the quality of the generated phonetic annotation. The architecture achieved a PER of 18.1 % on the entire test set, computed with all 101 unique phonetic symbols, showing slight differences between the individual languages. We conclude that Common Phone provides sufficient variability and reliable phonetic annotation to help bridging the gap between research and application of acoustic models.

Philipp Klumpp, Tomás Arias-Vergara, Juan Camilo Vásquez-Correa et al.

As one of the most prevalent neurodegenerative disorders, Parkinson's disease (PD) has a significant impact on the fine motor skills of patients. The complex interplay of different articulators during speech production and realization of required muscle tension become increasingly difficult, thus leading to a dysarthric speech. Characteristic patterns such as vowel instability, slurred pronunciation and slow speech can often be observed in the affected individuals and were analyzed in previous studies to determine the presence and progression of PD. In this work, we used a phonetic recognizer trained exclusively on healthy speech data to investigate how PD affected the phonetic footprint of patients. We rediscovered numerous patterns that had been described in previous contributions although our system had never seen any pathological speech previously. Furthermore, we could show that intermediate activations from the neural network could serve as feature vectors encoding information related to the disease state of individuals. We were also able to directly correlate the expert-rated intelligibility of a speaker with the mean confidence of phonetic predictions. Our results support the assumption that pathological data is not necessarily required to train systems that are capable of analyzing PD speech.

Luis Felipe Parra-Gallego, Juan Rafael Orozco-Arroyave

This paper focuses on finding suitable features to robustly recognize emotions and evaluate customer satisfaction from speech in real acoustic scenarios. The classification of emotions is based on standard and well-known corpora and the evaluation of customer satisfaction is based on recordings of real opinions given by customers about the received service during phone calls with call-center agents. The feature sets considered in this study include two speaker models, namely x-vectors and i-vectors, and also the well known feature set introduced in the Interspeech 2010 Paralinguistics Challenge (I2010PC). Additionally, we introduce the use of phonation, articulation and prosody features extracted with the DisVoice framework as alternative feature sets to robustly model emotions and customer satisfaction from speech. The results indicate that the I2010PC feature set is the best approach to classify emotions in the standard databases typically used in the literature. When considering the recordings collected in the call-center, without any control over the acoustic conditions, the best results are obtained with our articulation features. The I2010PC feature set includes 1584 measures while the articulation approach only includes 488 measures. We think that the proposed approach is more suitable for real-world applications where the acoustic conditions are not controlled and also it is potentially more convenient for industrial applications.

Daniel Escobar-Grisales, Juan Camilo Vasquez-Correa, Juan Rafael Orozco-Arroyave

The interest in demographic information retrieval based on text data has increased in the research community because applications have shown success in different sectors such as security, marketing, heath-care, and others. Recognition and identification of demographic traits such as gender, age, location, or personality based on text data can help to improve different marketing strategies. For instance it makes it possible to segment and to personalize offers, thus products and services are exposed to the group of greatest interest. This type of technology has been discussed widely in documents from social media. However, the methods have been poorly studied in data with a more formal structure, where there is no access to emoticons, mentions, and other linguistic phenomena that are only present in social media. This paper proposes the use of recurrent and convolutional neural networks, and a transfer learning strategy for gender recognition in documents that are written in informal and formal languages. Models are tested in two different databases consisting of Tweets and call-center conversations. Accuracies of up to 75\% are achieved for both databases. The results also indicate that it is possible to transfer the knowledge from a system trained on a specific type of expressions or idioms such as those typically used in social media into a more formal type of text data, where the amount of data is more scarce and its structure is completely different.

Luis Felipe Gomez-Gomez, Aythami Morales, Julian Fierrez et al.

Parkinson's Disease (PD) is a neurological disorder that affects facial movements and non-verbal communication. Patients with PD present a reduction in facial movements called hypomimia which is evaluated in item 3.2 of the MDS-UPDRS-III scale. In this work, we propose to use facial expression analysis from face images based on affective domains to improve PD detection. We propose different domain adaptation techniques to exploit the latest advances in face recognition and Face Action Unit (FAU) detection. The principal contributions of this work are: (1) a novel framework to exploit deep face architectures to model hypomimia in PD patients; (2) we experimentally compare PD detection based on single images vs. image sequences while the patients are evoked various face expressions; (3) we explore different domain adaptation techniques to exploit existing models initially trained either for Face Recognition or to detect FAUs for the automatic discrimination between PD patients and healthy subjects; and (4) a new approach to use triplet-loss learning to improve hypomimia modeling and PD detection. The results on real face images from PD patients show that we are able to properly model evoked emotions using image sequences (neutral, onset-transition, apex, offset-transition, and neutral) with accuracy improvements up to 5.5% (from 72.9% to 78.4%) with respect to single-image PD detection. We also show that our proposed affective-domain adaptation provides improvements in PD detection up to 8.9% (from 78.4% to 87.3% detection accuracy).