106 citations · 315 across the 24 of their papers we have counts for
29 papers
CARES: A Controlled Synthetic Benchmark of Speaker Reactions to Sound
Marcel Gibier, Thomas Thebaud, Olivier Boëffard +1
Automatic audio scene description turns a recording into a text account of a situation. One difficulty is deciding which elements of the audio should be kept, since a description c…
Acoustic domain shift in spoken language identification from systematic domain generalization evaluation to real-world application
Francois Derrida, Raphaël Duroselle, Thomas Courtat +1
Domain Generalization (DG) aims to develop models that remain robust to conditions unseen during training. While DG has been systematically studied in computer vision through contr…
Audio Question Answering with GRPO-Based Fine-Tuning and Calibrated Segment-Level Predictions
Marcel Gibier, Nolwenn Celton, Raphaël Duroselle +3
In this report, we describe our submission to Track 5 of the DCASE 2025 Challenge for the task of Audio Question Answering(AQA). Our system leverages the SSL backbone BEATs to extr…
Segmentwise Pruning in Audio-Language Models
Marcel Gibier, Raphaël Duroselle, Pierre Serrano +2
Recent audio-language models have shown impressive performance across a wide range of audio tasks and are increasingly capable of handling long audio inputs. However, the computing…
Improving Out-of-Domain Audio Deepfake Detection via Layer Selection and Fusion of SSL-Based Countermeasures
Pierre Serrano, Raphaël Duroselle, Florian Angulo +2
Audio deepfake detection systems based on frozen pre-trained self-supervised learning (SSL) encoders show a high level of performance when combined with layer-weighted pooling meth…
The distribution of calibrated likelihood functions on the probability-likelihood Aitchison simplex
Paul-Gauthier Noé, Andreas Nautsch, Driss Matrouf +2
While calibration of probabilistic predictions has been widely studied, this paper rather addresses calibration of likelihood functions. This has been discussed, especially in biom…
Explaining a probabilistic prediction on the simplex with Shapley compositions
Paul-Gauthier Noé, Miquel Perelló-Nieto, Jean-François Bonastre +1
Originating in game theory, Shapley values are widely used for explaining a machine learning model's prediction by quantifying the contribution of each feature's value to the predi…
a-DCF: an architecture agnostic metric with application to spoofing-robust speaker verification
Hye-jin Shim, Jee-weon Jung, Tomi Kinnunen +3
Spoofing detection is today a mainstream research topic. Standard metrics can be applied to evaluate the performance of isolated spoofing detection solutions and others have been p…
Effect of utterance duration and phonetic content on speaker identification using second-order statistical methods
Ivan Magrin-Chagnolleau, Jean François Bonastre, Frédéric Bimbot
Second-order statistical methods show very good results for automatic speaker identification in controlled recording conditions. These approaches are generally used on the entire s…
Thech. Report: Genuinization of Speech waveform PMF for speaker detection spoofing and countermeasures
Itshak Lapidot, Jean-Francois Bonastre
In the context of spoofing attacks in speaker recognition systems, we observed that the waveform probability mass function (PMF) of genuine speech differs significantly from the PM…
SynVox2: Towards a privacy-friendly VoxCeleb2 dataset
Xiaoxiao Miao, Xin Wang, Erica Cooper +5
The success of deep learning in speaker recognition relies heavily on the use of large datasets. However, the data-hungry nature of deep learning methods has already being question…
Federated Learning for ASR based on Wav2vec 2.0
Tuan Nguyen, Salima Mdhaffar, Natalia Tomashenko +2
This paper presents a study on the use of federated learning to train an ASR model based on a wav2vec 2.0 model pre-trained by self supervision. Carried out on the well-known TED-L…
Hiding speaker's sex in speech using zero-evidence speaker representation in an analysis/synthesis pipeline
Paul-Gauthier Noé, Xiaoxiao Miao, Xin Wang +3
The use of modern vocoders in an analysis/synthesis pipeline allows us to investigate high-quality voice conversion that can be used for privacy purposes. Here, we propose to trans…
The VoicePrivacy 2020 Challenge Evaluation Plan
Natalia Tomashenko, Brij Mohan Lal Srivastava, Xin Wang +8
The VoicePrivacy Challenge aims to promote the development of privacy preservation tools for speech technology by gathering a new community to define the tasks of interest and the…
The VoicePrivacy 2022 Challenge Evaluation Plan
Natalia Tomashenko, Xin Wang, Xiaoxiao Miao +7
For new participants - Executive summary: (1) The task is to develop a voice anonymization system for speech data which conceals the speaker's voice identity while protecting lingu…
Retrieving Speaker Information from Personalized Acoustic Models for Speech Recognition
Salima Mdhaffar, Jean-François Bonastre, Marc Tommasi +2
The widespread of powerful personal devices capable of collecting voice of their users has opened the opportunity to build speaker adapted speech recognition system (ASR) or to par…
Privacy attacks for automatic speech recognition acoustic models in a federated learning framework
Natalia Tomashenko, Salima Mdhaffar, Marc Tommasi +2
This paper investigates methods to effectively retrieve speaker information from the personalized speaker adapted neural network acoustic models (AMs) in automatic speech recogniti…
Revisiting Speech Content Privacy
Jennifer Williams, Junichi Yamagishi, Paul-Gauthier Noe +2
In this paper, we discuss an important aspect of speech privacy: protecting spoken content. New capabilities from the field of machine learning provide a unique and timely opportun…
A bridge between features and evidence for binary attribute-driven perfect privacy
Paul-Gauthier Noé, Andreas Nautsch, Driss Matrouf +2
Attribute-driven privacy aims to conceal a single user's attribute, contrary to anonymisation that tries to hide the full identity of the user in some data. When the attribute to p…
The VoicePrivacy 2020 Challenge: Results and findings
Natalia Tomashenko, Xin Wang, Emmanuel Vincent +11
This paper presents the results and analyses stemming from the first VoicePrivacy 2020 Challenge which focuses on developing anonymization solutions for speech technology. We provi…
Benchmarking and challenges in security and privacy for voice biometrics
Jean-Francois Bonastre, Hector Delgado, Nicholas Evans +13
For many decades, research in speech technologies has focused upon improving reliability. With this now meeting user expectations for a range of diverse applications, speech techno…
Adversarial Disentanglement of Speaker Representation for Attribute-Driven Privacy Preservation
Paul-Gauthier Noé, Mohammad Mohammadamini, Driss Matrouf +3
In speech technologies, speaker's voice representation is used in many applications such as speech recognition, voice conversion, speech synthesis and, obviously, user authenticati…
Speech Pseudonymisation Assessment Using Voice Similarity Matrices
Paul-Gauthier Noé, Jean-François Bonastre, Driss Matrouf +3
The proliferation of speech technologies and rising privacy legislation calls for the development of privacy preservation solutions for speech applications. These are essential sin…
The Privacy ZEBRA: Zero Evidence Biometric Recognition Assessment
Andreas Nautsch, Jose Patino, Natalia Tomashenko +5
Mounting privacy legislation calls for the preservation of privacy in speech technology, though solutions are gravely lacking. While evaluation campaigns are long-proven tools to d…