Picture for Yannick Estève

Yannick Estève

LIA

Open Implementation and Study of BEST-RQ for Speech Processing

Add code
May 07, 2024
Viaarxiv icon

Is one brick enough to break the wall of spoken dialogue state tracking?

Nov 03, 2023
Viaarxiv icon

Enhancing expressivity transfer in textless speech-to-speech translation

Oct 11, 2023
Viaarxiv icon

Acoustic and linguistic representations for speech continuous emotion recognition in call center conversations

Oct 06, 2023
Figure 1 for Acoustic and linguistic representations for speech continuous emotion recognition in call center conversations
Figure 2 for Acoustic and linguistic representations for speech continuous emotion recognition in call center conversations
Figure 3 for Acoustic and linguistic representations for speech continuous emotion recognition in call center conversations
Figure 4 for Acoustic and linguistic representations for speech continuous emotion recognition in call center conversations
Viaarxiv icon

Semantic enrichment towards efficient speech representations

Jul 03, 2023
Figure 1 for Semantic enrichment towards efficient speech representations
Figure 2 for Semantic enrichment towards efficient speech representations
Figure 3 for Semantic enrichment towards efficient speech representations
Figure 4 for Semantic enrichment towards efficient speech representations
Viaarxiv icon

Learning Multilingual Expressive Speech Representation for Prosody Prediction without Parallel Data

Jun 29, 2023
Figure 1 for Learning Multilingual Expressive Speech Representation for Prosody Prediction without Parallel Data
Figure 2 for Learning Multilingual Expressive Speech Representation for Prosody Prediction without Parallel Data
Figure 3 for Learning Multilingual Expressive Speech Representation for Prosody Prediction without Parallel Data
Figure 4 for Learning Multilingual Expressive Speech Representation for Prosody Prediction without Parallel Data
Viaarxiv icon

Some voices are too common: Building fair speech recognition systems using the Common Voice dataset

Jun 01, 2023
Figure 1 for Some voices are too common: Building fair speech recognition systems using the Common Voice dataset
Figure 2 for Some voices are too common: Building fair speech recognition systems using the Common Voice dataset
Figure 3 for Some voices are too common: Building fair speech recognition systems using the Common Voice dataset
Figure 4 for Some voices are too common: Building fair speech recognition systems using the Common Voice dataset
Viaarxiv icon

OLISIA: a Cascade System for Spoken Dialogue State Tracking

Add code
Apr 20, 2023
Figure 1 for OLISIA: a Cascade System for Spoken Dialogue State Tracking
Figure 2 for OLISIA: a Cascade System for Spoken Dialogue State Tracking
Figure 3 for OLISIA: a Cascade System for Spoken Dialogue State Tracking
Figure 4 for OLISIA: a Cascade System for Spoken Dialogue State Tracking
Viaarxiv icon

Improving Accented Speech Recognition with Multi-Domain Training

Mar 14, 2023
Figure 1 for Improving Accented Speech Recognition with Multi-Domain Training
Figure 2 for Improving Accented Speech Recognition with Multi-Domain Training
Figure 3 for Improving Accented Speech Recognition with Multi-Domain Training
Figure 4 for Improving Accented Speech Recognition with Multi-Domain Training
Viaarxiv icon

Federated Learning for ASR based on Wav2vec 2.0

Add code
Feb 20, 2023
Figure 1 for Federated Learning for ASR based on Wav2vec 2.0
Figure 2 for Federated Learning for ASR based on Wav2vec 2.0
Figure 3 for Federated Learning for ASR based on Wav2vec 2.0
Figure 4 for Federated Learning for ASR based on Wav2vec 2.0
Viaarxiv icon