[PDF][PDF] x-Vectors Meet Adversarial Attacks: Benchmarking Adversarial Robustness in Speaker Verification.

J Villalba, Y Zhang, N Dehak - Interspeech, 2020 - interspeech2020.org
Abstract Automatic Speaker Verification (ASV) enables high-security applications like user
authentication or criminal investigation. However, ASV can be subjected to malicious …

[PDF][PDF] Black-Box Attacks on Spoofing Countermeasures Using Transferability of Adversarial Examples.

Y Zhang, Z Jiang, J Villalba, N Dehak - Interspeech, 2020 - isca-archive.org
Spoofing countermeasure systems protect Automatic Speaker Verification (ASV) systems
from spoofing attacks such as replay, synthesis, and conversion. However, research has …

Robust speaker recognition with transformers using wav2vec 2.0

S Novoselov, G Lavrentyeva, A Avdeeva… - arxiv preprint arxiv …, 2022 - arxiv.org
Recent advances in unsupervised speech representation learning discover new
approaches and provide new state-of-the-art for diverse types of speech processing tasks …

[PDF][PDF] Spine2Net: SpineNet with Res2Net and Time-Squeeze-and-Excitation Blocks for Speaker Recognition.

M Rybicka, J Villalba, P Zelasko, N Dehak… - Interspeech, 2021 - isca-archive.org
Modeling speaker embeddings using deep neural networks is currently state-of-the-art in
speaker recognition. Recently, ResNet-based structures have gained a broader interest …

[PDF][PDF] STC Speaker Recognition System for the NIST SRE 2021.

G Lavrentyeva, S Novoselov, V Volokhov, A Avdeeva… - Odyssey, 2022 - isca-archive.org
Abstract The 2021 Speaker Recognition Evaluation (SRE21) is the next of an open speaker
recognition evaluations conducted by the US National Institute of Standards and Technology …

[PDF][PDF] On the robustness of wav2vec 2.0 based speaker recognition systems

S Novoselov, G Lavrentyeva, A Avdeeva… - Proc …, 2023 - isca-archive.org
Recent advances in unsupervised speech representation learning discover new
approaches and provide new state-of-the-art for diverse types of speech processing tasks …

Representation learning to classify and detect adversarial attacks against speaker and speech recognition systems

J Villalba, S Joshi, P Żelasko, N Dehak - arxiv preprint arxiv:2107.04448, 2021 - arxiv.org
Adversarial attacks have become a major threat for machine learning applications. There is
a growing interest in studying these attacks in the audio domain, eg, speech and speaker …

[PDF][PDF] Advances in Cross-Lingual and Cross-Source Audio-Visual Speaker Recognition: The JHU-MIT System for NIST SRE21.

J Villalba, BJ Borgstrom, S Kataria, M Rybicka… - Odyssey, 2022 - researchgate.net
We present a condensed description of the joint effort of JHUCLSP/HLTCOE, MIT-LL and
AGH for NIST SRE21. NIST SRE21 consisted of speaker detection over multilingual …

Deep feature cyclegans: Speaker identity preserving non-parallel microphone-telephone domain adaptation for speaker verification

S Kataria, J Villalba, P Żelasko… - arxiv preprint arxiv …, 2021 - arxiv.org
With the increase in the availability of speech from varied domains, it is imperative to use
such out-of-domain data to improve existing speech systems. Domain adaptation is a …

Time-domain speech super-resolution with gan based modeling for telephony speaker verification

S Kataria, J Villalba, L Moro-Velázquez… - … on Audio, Speech …, 2024 - ieeexplore.ieee.org
Automatic Speaker Verification (ASV) technology has become commonplace in virtual
assistants. However, its performance suffers when there is a mismatch between the train and …