Picture for Hervé Bredin

Hervé Bredin

IRIT-SAMoVA

PixIT: Joint Training of Speaker Diarization and Speech Separation from Real-world Multi-speaker Recordings

Add code
Mar 04, 2024
Figure 1 for PixIT: Joint Training of Speaker Diarization and Speech Separation from Real-world Multi-speaker Recordings
Figure 2 for PixIT: Joint Training of Speaker Diarization and Speech Separation from Real-world Multi-speaker Recordings
Figure 3 for PixIT: Joint Training of Speaker Diarization and Speech Separation from Real-world Multi-speaker Recordings
Figure 4 for PixIT: Joint Training of Speaker Diarization and Speech Separation from Real-world Multi-speaker Recordings
Viaarxiv icon

Powerset multi-class cross entropy loss for neural speaker diarization

Add code
Oct 19, 2023
Viaarxiv icon

BabySLM: language-acquisition-friendly benchmark of self-supervised spoken language models

Add code
Jun 08, 2023
Figure 1 for BabySLM: language-acquisition-friendly benchmark of self-supervised spoken language models
Figure 2 for BabySLM: language-acquisition-friendly benchmark of self-supervised spoken language models
Figure 3 for BabySLM: language-acquisition-friendly benchmark of self-supervised spoken language models
Figure 4 for BabySLM: language-acquisition-friendly benchmark of self-supervised spoken language models
Viaarxiv icon

Brouhaha: multi-task training for voice activity detection, speech-to-noise ratio, and C50 room acoustics estimation

Add code
Oct 27, 2022
Figure 1 for Brouhaha: multi-task training for voice activity detection, speech-to-noise ratio, and C50 room acoustics estimation
Figure 2 for Brouhaha: multi-task training for voice activity detection, speech-to-noise ratio, and C50 room acoustics estimation
Figure 3 for Brouhaha: multi-task training for voice activity detection, speech-to-noise ratio, and C50 room acoustics estimation
Figure 4 for Brouhaha: multi-task training for voice activity detection, speech-to-noise ratio, and C50 room acoustics estimation
Viaarxiv icon

Overlap-aware low-latency online speaker diarization based on end-to-end local segmentation

Add code
Sep 14, 2021
Figure 1 for Overlap-aware low-latency online speaker diarization based on end-to-end local segmentation
Figure 2 for Overlap-aware low-latency online speaker diarization based on end-to-end local segmentation
Figure 3 for Overlap-aware low-latency online speaker diarization based on end-to-end local segmentation
Figure 4 for Overlap-aware low-latency online speaker diarization based on end-to-end local segmentation
Viaarxiv icon

End-to-end speaker segmentation for overlap-aware resegmentation

Apr 08, 2021
Figure 1 for End-to-end speaker segmentation for overlap-aware resegmentation
Figure 2 for End-to-end speaker segmentation for overlap-aware resegmentation
Figure 3 for End-to-end speaker segmentation for overlap-aware resegmentation
Figure 4 for End-to-end speaker segmentation for overlap-aware resegmentation
Viaarxiv icon

A Comparison of Metric Learning Loss Functions for End-To-End Speaker Verification

Add code
Mar 31, 2020
Figure 1 for A Comparison of Metric Learning Loss Functions for End-To-End Speaker Verification
Figure 2 for A Comparison of Metric Learning Loss Functions for End-To-End Speaker Verification
Figure 3 for A Comparison of Metric Learning Loss Functions for End-To-End Speaker Verification
Figure 4 for A Comparison of Metric Learning Loss Functions for End-To-End Speaker Verification
Viaarxiv icon

The Speed Submission to DIHARD II: Contributions & Lessons Learned

Nov 06, 2019
Figure 1 for The Speed Submission to DIHARD II: Contributions & Lessons Learned
Figure 2 for The Speed Submission to DIHARD II: Contributions & Lessons Learned
Figure 3 for The Speed Submission to DIHARD II: Contributions & Lessons Learned
Figure 4 for The Speed Submission to DIHARD II: Contributions & Lessons Learned
Viaarxiv icon

LSTM based Similarity Measurement with Spectral Clustering for Speaker Diarization

Add code
Jul 23, 2019
Figure 1 for LSTM based Similarity Measurement with Spectral Clustering for Speaker Diarization
Figure 2 for LSTM based Similarity Measurement with Spectral Clustering for Speaker Diarization
Figure 3 for LSTM based Similarity Measurement with Spectral Clustering for Speaker Diarization
Figure 4 for LSTM based Similarity Measurement with Spectral Clustering for Speaker Diarization
Viaarxiv icon

TristouNet: Triplet Loss for Speaker Turn Embedding

Add code
Apr 11, 2017
Figure 1 for TristouNet: Triplet Loss for Speaker Turn Embedding
Figure 2 for TristouNet: Triplet Loss for Speaker Turn Embedding
Figure 3 for TristouNet: Triplet Loss for Speaker Turn Embedding
Figure 4 for TristouNet: Triplet Loss for Speaker Turn Embedding
Viaarxiv icon