Search

Your search keyword '"Md. Sahidullah"' showing total 6 results
Start Over Author "Md. Sahidullah" Publisher hal ccsd
6 results on '"Md. Sahidullah"'

1. Advancing Stuttering Detection via Data Augmentation, Class-Balanced Loss and Multi-Contextual Deep Learning

Author

Shakeel A. Sheikh, Md Sahidullah, Fabrice Hirsch, Slim Ouni, Speech Modeling for Facilitating Oral-Based Communication (MULTISPEECH), Inria Nancy - Grand Est, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Department of Natural Language Processing & Knowledge Discovery (LORIA - NLPKD), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS), Praxiling (Praxiling), and Université Paul-Valéry - Montpellier 3 (UPVM)-Centre National de la Recherche Scientifique (CNRS)

Subjects
FOS: Computer and information sciences, Sound (cs.SD), Computer Science - Machine Learning, Data augmentation, Stuttering detection, Health Informatics, Class balanced learning, Computer Science - Sound, Computer Science Applications, Machine Learning (cs.LG), [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Health Information Management, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Audio and Speech Processing (eess.AS), FOS: Electrical engineering, electronic engineering, information engineering, Electrical and Electronic Engineering, Speech disorder, Electrical Engineering and Systems Science - Audio and Speech Processing
Abstract

Stuttering is a neuro-developmental speech impairment characterized by uncontrolled utterances (interjections) and core behaviors (blocks, repetitions, and prolongations), and is caused by the failure of speech sensorimotors. Due to its complex nature, stuttering detection (SD) is a difficult task. If detected at an early stage, it could facilitate speech therapists to observe and rectify the speech patterns of persons who stutter (PWS). The stuttered speech of PWS is usually available in limited amounts and is highly imbalanced. To this end, we address the class imbalance problem in the SD domain via a multibranching (MB) scheme and by weighting the contribution of classes in the overall loss function, resulting in a huge improvement in stuttering classes on the SEP-28k dataset over the baseline (StutterNet). To tackle data scarcity, we investigate the effectiveness of data augmentation on top of a multi-branched training scheme. The augmented training outperforms the MB StutterNet (clean) by a relative margin of 4.18% in macro F1-score (F1). In addition, we propose a multi-contextual (MC) StutterNet, which exploits different contexts of the stuttered speech, resulting in an overall improvement of 4.48% in F 1 over the single context based MB StutterNet. Finally, we have shown that applying data augmentation in the cross-corpora scenario can improve the overall SD performance by a relative margin of 13.23% in F1 over the clean training., Accepted in IEEE Journal of Biomedical Health Informatics 2023

Published
2023
Full Text
View/download PDF

2. Robust acoustic domain identification with its application to speaker diarization

Author

A Kishore Kumar, Shefali Waldekar, Md Sahidullah, Goutam Saha, Indian Institute of Technology Kharagpur (IIT Kharagpur), GITAM School of Technology, Speech Modeling for Facilitating Oral-Based Communication (MULTISPEECH), Inria Nancy - Grand Est, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Department of Natural Language Processing & Knowledge Discovery (LORIA - NLPKD), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS), Jagadish Chandra Bose Centre of Advanced Technology (JCBCAT), and Grid5000

Subjects
FOS: Computer and information sciences, Acoustic domain identification (ADI), Diarization error rate (DER), Linguistics and Language, Sound (cs.SD), [INFO.INFO-MM]Computer Science [cs]/Multimedia [cs.MM], [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], Language and Linguistics, Computer Science - Sound, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Human-Computer Interaction, [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, Audio and Speech Processing (eess.AS), Agglomerative hierarchical clustering (AHC), Speaker diarization, [INFO.INFO-SD]Computer Science [cs]/Sound [cs.SD], FOS: Electrical engineering, electronic engineering, information engineering, Computer Vision and Pattern Recognition, Jaccard error rate (JER), Software, Electrical Engineering and Systems Science - Audio and Speech Processing
Abstract

With the rise in multimedia content over the years, more variety is observed in the recording environments of audio. An audio processing system might benefit when it has a module to identify the acoustic domain at its front-end. In this paper, we demonstrate the idea of \emph{acoustic domain identification} (ADI) for \emph{speaker diarization}. For this, we first present a detailed study of the various domains of the third DIHARD challenge highlighting the factors that differentiated them from each other. Our main contribution is to develop a simple and efficient solution for ADI. In the present work, we explore speaker embeddings for this task. Next, we integrate the ADI module with the speaker diarization framework of the DIHARD III challenge. The performance substantially improved over that of the baseline when the thresholds for agglomerative hierarchical clustering were optimized according to the respective domains. We achieved a relative improvement of more than $5\%$ and $8\%$ in DER for core and full conditions, respectively, on Track 1 of the DIHARD III evaluation set., Accepted for publication in International Journal of Speech Technology (Springer Nature)

Published
2022
Full Text
View/download PDF

3. Privacy and utility of x-vector based speaker anonymization

Author

Brij Mohan Lal Srivastava, Mohamed Maouche, Md Sahidullah, Emmanuel Vincent, Aurelien Bellet, Marc Tommasi, Natalia Tomashenko, Xin Wang, Junichi Yamagishi, Machine Learning in Information Networks (MAGNET), Inria Lille - Nord Europe, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 (CRIStAL), Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS)-Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS), Speech Modeling for Facilitating Oral-Based Communication (MULTISPEECH), Inria Nancy - Grand Est, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Department of Natural Language Processing & Knowledge Discovery (LORIA - NLPKD), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS), Laboratoire Informatique d'Avignon (LIA), Avignon Université (AU)-Centre d'Enseignement et de Recherche en Informatique - CERI, National Institute of Informatics (NII), Grid'5000, ANR-18-CE23-0018,DEEP-PRIVACY,Apprentissage distribué, personnalisé, préservant la privacité pour le traitement de la parole(2018), ANR-19-DATA-0008,Harpocrates,Open data, outils et challenges pour l'anonymisation des voix(2019), European Project: 825081,H2020,COMPRISE(2018), Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL), Centre d'Enseignement et de Recherche en Informatique - CERI-Avignon Université (AU), Machine Learning in Information Networks [MAGNET], Speech Modeling for Facilitating Oral-Based Communication [MULTISPEECH], Laboratoire Informatique d'Avignon [LIA], National Institute of Informatics [NII], Srivastava, Brij Mohan Lal, APPEL À PROJETS GÉNÉRIQUE 2018 - Apprentissage distribué, personnalisé, préservant la privacité pour le traitement de la parole - - DEEP-PRIVACY2018 - ANR-18-CE23-0018 - AAPG2018 - VALID, Open data, outils et challenges pour l'anonymisation des voix - - Harpocrates2019 - ANR-19-DATA-0008 - DONNEES - VALID, and Cost effective, Multilingual, Privacy-driven voice-enabled Services - COMPRISE - - H20202018-12-01 - 2021-11-30 - 825081 - VALID

Subjects
[INFO.INFO-AI] Computer Science [cs]/Artificial Intelligence [cs.AI], speaker anonymization, speaker identification, Acoustics and Ultrasonics, speech recognition, privacy, linkability, utility, [INFO.INFO-LG] Computer Science [cs]/Machine Learning [cs.LG], [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL], [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Computational Mathematics, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], [INFO.INFO-CL] Computer Science [cs]/Computation and Language [cs.CL], Computer Science (miscellaneous), Electrical and Electronic Engineering
Abstract

International audience; We study the scenario where individuals (speakers) contribute to the publication of an anonymized speech corpus. Data users then leverage this public corpus to perform downstream tasks (such as training automatic speech recognition systems), while attackers may try to de-anonymize itbased on auxiliary knowledge they collect. Motivated by this scenario, speaker anonymization aims to conceal the speaker identity while preserving the quality and usefulness of speech data. In this paper, we study x-vector based speaker anonymization, the leading approach in the recent Voice Privacy Challenge, which converts an input utterance into that of a random pseudo-speaker. We show that the strength of the anonymization varies significantly depending on how the pseudo-speaker is selected. In particular, we investigate four design choices: the distance measure between speakers, the region of x-vector space where the pseudo-speaker is mapped, the gender selection and whether to use speaker or utterance level assignment. We assess the quality of anonymization from the perspective of the three actors involved in our threat model, namely the speaker, the user and the attacker. To measure privacy and utility, we use respectively the linkability score achieved by the attackers and the decoding word error rate incurred by an ASR model trained with the anonymized data. Experiments on LibriSpeech dataset confirm that the optimal combination ofdesign choices yield state-of-the-art performance in terms of privacy protection as well as utility. Experiments on Mozilla Common Voice dataset show that the best design choices with 50 speakers guarantee the same anonymization level against re-identification attack as raw speech with 20,000 speakers.

Published
2022
Full Text
View/download PDF

4. Modeling and training strategies for language recognition systems

Author

Raphaël Duroselle, Denis Jouvet, Md. Sahidullah, Irina Illina, Speech Modeling for Facilitating Oral-Based Communication (MULTISPEECH), Inria Nancy - Grand Est, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Department of Natural Language Processing & Knowledge Discovery (LORIA - NLPKD), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS), Experiments presented in this paper were carried out using the Grid5000 testbed, supported by a scientific interest group hosted by Inria and including CNRS, RENATER and several Universities as well as other organizations (see https://www.grid5000.fr). This work has been partly funded by the French Direction Générale de l’Armement., Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), and Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)

Subjects
business.industry, Computer science, multi-task training, 020206 networking & telecommunications, 02 engineering and technology, computer.software_genre, endto-end speech recognition, Training (civil), language recognition, 030507 speech-language pathology & audiology, 03 medical and health sciences, bottleneck features, 0202 electrical engineering, electronic engineering, information engineering, [INFO]Computer Science [cs], Artificial intelligence, 0305 other medical science, business, computer, Natural language processing, Language recognition
Abstract

International audience; Automatic speech recognition is complementary to language recognition. The language recognition systems exploit this complementarity by using frame-level bottleneck features extracted from neural networks trained with a phone recognition task. Recent methods apply frame-level bottleneck features extracted from an end-to-end sequence-to-sequence speech recognition model. In this work, we study an integrated approach of the training of the speech recognition feature extractor and language recognition modules. We show that for both classical phone recognition and end-to-end sequence-to-sequence features, sequential training of the two modules is not the optimal strategy. The feature extractor can be improved by supervision with the language identification loss, either in a fine-tuning step or in a multi-task training framework. Besides, we notice that end-to-end sequence-to-sequence bottleneck features are on par with classical phone recognition bottleneck features without requiring a forced alignment of the signal with target tokens. However, for sequence-to-sequence, the architecture of the model seems to play an important role; the Conformer architectures leads to much better results than the conventional stacked DNNs approach; and can even be trained directly with the LID module in an end-to-end approach.

Published
2021
Full Text
View/download PDF

5. Cross-Corpora Language Recognition: A Preliminary Investigation with Indian Languages

Author

Spandan Dey, Goutam Saha, Md Sahidullah, Indian Institute of Technology Kharagpur (IIT Kharagpur), Speech Modeling for Facilitating Oral-Based Communication (MULTISPEECH), Inria Nancy - Grand Est, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Department of Natural Language Processing & Knowledge Discovery (LORIA - NLPKD), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), and Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)

Subjects
FOS: Computer and information sciences, TDNN, [SPI.ACOU]Engineering Sciences [physics]/Acoustics [physics.class-ph], Sound (cs.SD), long-term average spectrum, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], 02 engineering and technology, Computer Science - Sound, language recognition, 030507 speech-language pathology & audiology, 03 medical and health sciences, Cross-corpora, Audio and Speech Processing (eess.AS), 0202 electrical engineering, electronic engineering, information engineering, FOS: Electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, 0305 other medical science, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, channel compensation, Electrical Engineering and Systems Science - Audio and Speech Processing
Abstract

In this paper, we conduct one of the very first studies for cross-corpora performance evaluation in the spoken language identification (LID) problem. Cross-corpora evaluation was not explored much in LID research, especially for the Indian languages. We have selected three Indian spoken language corpora: IIITH-ILSC, LDC South Asian, and IITKGP-MLILSC. For each of the corpus, LID systems are trained on the state-of-the-art time-delay neural network (TDNN) based architecture with MFCC features. We observe that the LID performance degrades drastically for cross-corpora evaluation. For example, the system trained on the IIITH-ILSC corpus shows an average EER of 11.80 % and 43.34 % when evaluated with the same corpora and LDC South Asian corpora, respectively. Our preliminary analysis shows the significant differences among these corpora in terms of mismatch in the long-term average spectrum (LTAS) and signal-to-noise ratio (SNR). Subsequently, we apply different feature level compensation methods to reduce the cross-corpora acoustic mismatch. Our results indicate that these feature normalization schemes can help to achieve promising LID performance on cross-corpora experiments., Comment: Accepted in EUSIPCO 2021 : European Signal Processing Conference

Published
2021
Full Text
View/download PDF

6. Speech Frame Selection for Spoofing Detection with an Application to Partially Spoofed Audio-Data

Author

Goutam Saha, Sahidullah, A Kishore Kumar, Monisankha Pal, Dipjyoti Paul, Indian Institute of Technology Kharagpur (IIT Kharagpur), University of Crete [Heraklion] (UOC), University of Southern California (USC), Speech Modeling for Facilitating Oral-Based Communication (MULTISPEECH), Inria Nancy - Grand Est, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Department of Natural Language Processing & Knowledge Discovery (LORIA - NLPKD), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS), We would like acknowledge the funding agencies. Dipjyoti Paul’s work is funded by the EUs H2020 research and innovation programme under theMSCA GA 67532 (the ENRICH network: www.enrich-etn.eu). The work of Md Sahidullah is supported by Region Grand Est., Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), and Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)

Subjects
Linguistics and Language, Spoofing attack, Computer science, Speech recognition, Feature extraction, Selection strategy, Anti-spoofing, Frame selection, Language and Linguistics, Synthetic speech detection, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], 030507 speech-language pathology & audiology, 03 medical and health sciences, Countermeasures, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], ASVspoof, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], Mixture model, Human-Computer Interaction, Partially spoofed speech, Speaker verification, Computer Vision and Pattern Recognition, Mel-frequency cepstrum, Performance improvement, 0305 other medical science, Classifier (UML), [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, Software, Utterance, Voice conversion
Abstract

International audience; In this paper, we introduce a frame selection strategy for improved detection of spoofed speech. A countermeasure (CM) system typically uses a Gaussian mixture model (GMM) based classifier for computing the log-likelihood scores. The average log-likelihood ratio for all speech frames of a test utterance is calculated as the score for the decision making. As opposed to this standard approach, we propose to use selected speech frames of the test utterance for scoring. We present two simple and computationally efficient frame selection strategies based on the log-likelihood ratios of the individual frames. The performance is evaluated with constant-Q cepstral coefficients as front-end feature extraction and two-class GMM as a back-end classifier. We conduct the experiments using the speech corpora from ASVspoof 2015, 2017, and 2019 challenges. The experimental results show that the proposed scoring techniques substantially outperform the conventional scoring technique for both the development and evaluation data set of ASVspoof 2015 corpus. We did not observe noticeable performance gain in ASVspoof 2017 and ASVspoof 2019 corpus. We further conducted experiments with partially spoofed data where spoofed data is created by augmenting natural and spoofed speech. In this scenario, the proposed methods demonstrate considerable performance improvement over baseline.

Published
2021
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results