Výsledky vyhledávání

Report

Exploring WavLM Back-ends for Speech Spoofing and Deepfake Detection

Autor: Stourbe, Theophile, Miara, Victor, Lepage, Theo, Dehak, Reda

This paper describes our submitted systems to the ASVspoof 5 Challenge Track 1: Speech Deepfake Detection - Open Condition, which consists of a stand-alone speech deepfake (bonafide vs spoof) detection task. Recently, large-scale self-supervised mode

Externí odkaz: http://arxiv.org/abs/2409.05032

Zobrazit plný text záznamu

Report

InaGVAD : a Challenging French TV and Radio Corpus Annotated for Speech Activity Detection and Speaker Gender Segmentation

Autor: Doukhan, David, Maertens, Christine, Personnic, William Le, Speroni, Ludovic, Dehak, Reda

Publikováno v: Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 8963-8974, Torino, Italia. ELRA and ICCL

InaGVAD is an audio corpus collected from 10 French radio and 18 TV channels categorized into 4 groups: generalist radio, music radio, news TV, and generalist TV. It contains 277 1-minute-long annotated recordings aimed at representing the acoustic d

Externí odkaz: http://arxiv.org/abs/2406.04429

Zobrazit plný text záznamu

Report

Towards Supervised Performance on Speaker Verification with Self-Supervised Learning by Leveraging Large-Scale ASR Models

Autor: Miara, Victor, Lepage, Theo, Dehak, Reda

Recent advancements in Self-Supervised Learning (SSL) have shown promising results in Speaker Verification (SV). However, narrowing the performance gap with supervised systems remains an ongoing challenge. Several studies have observed that speech re

Externí odkaz: http://arxiv.org/abs/2406.02285

Zobrazit plný text záznamu

Report

Additive Margin in Contrastive Self-Supervised Frameworks to Learn Discriminative Speaker Representations

Autor: Lepage, Theo, Dehak, Reda

Self-Supervised Learning (SSL) frameworks became the standard for learning robust class representations by benefiting from large unlabeled datasets. For Speaker Verification (SV), most SSL systems rely on contrastive-based loss functions. We explore

Externí odkaz: http://arxiv.org/abs/2404.14913

Zobrazit plný text záznamu

Report

Experimenting with Additive Margins for Contrastive Self-Supervised Speaker Verification

Autor: Lepage, Theo, Dehak, Reda

Publikováno v: Proc. INTERSPEECH 2023

Most state-of-the-art self-supervised speaker verification systems rely on a contrastive-based objective function to learn speaker representations from unlabeled speech data. We explore different ways to improve the performance of these methods by: (

Externí odkaz: http://arxiv.org/abs/2306.03664

Zobrazit plný text záznamu

Report

A benchmark for toxic comment classification on Civil Comments dataset

Autor: Duchene, Corentin, Jamet, Henri, Guillaume, Pierre, Dehak, Reda

Publikováno v: EGC 2023, vol. RNTI-E-39, pp.19-30

Toxic comment detection on social media has proven to be essential for content moderation. This paper compares a wide set of different models on a highly skewed multi-label hate speech dataset. We consider inference time and several metrics to measur

Externí odkaz: http://arxiv.org/abs/2301.11125

Zobrazit plný text záznamu

Report

Label-Efficient Self-Supervised Speaker Verification With Information Maximization and Contrastive Learning

Autor: Lepage, Théo, Dehak, Réda

Publikováno v: INTERSPEECH 2022, September 18-22, Incheon, Korea

State-of-the-art speaker verification systems are inherently dependent on some kind of human supervision as they are trained on massive amounts of labeled data. However, manually annotating utterances is slow, expensive and not scalable to the amount

Externí odkaz: http://arxiv.org/abs/2207.05506

Zobrazit plný text záznamu