AV2Wav: Diffusion-Based Re-synthesis from Continuous Self-supervised Features for Audio-Visual Speech Enhancement

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chou, Ju-Chieh, Chien, Chung-Ming, Livescu, Karen
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916466491654144
author Chou, Ju-Chieh
Chien, Chung-Ming
Livescu, Karen
author_facet Chou, Ju-Chieh
Chien, Chung-Ming
Livescu, Karen
contents Speech enhancement systems are typically trained using pairs of clean and noisy speech. In audio-visual speech enhancement (AVSE), there is not as much ground-truth clean data available; most audio-visual datasets are collected in real-world environments with background noise and reverberation, hampering the development of AVSE. In this work, we introduce AV2Wav, a resynthesis-based audio-visual speech enhancement approach that can generate clean speech despite the challenges of real-world training data. We obtain a subset of nearly clean speech from an audio-visual corpus using a neural quality estimator, and then train a diffusion model on this subset to generate waveforms conditioned on continuous speech representations from AV-HuBERT with noise-robust training. We use continuous rather than discrete representations to retain prosody and speaker information. With this vocoding task alone, the model can perform speech enhancement better than a masking-based baseline. We further fine-tune the diffusion model on clean/noisy utterance pairs to improve the performance. Our approach outperforms a masking-based baseline in terms of both automatic metrics and a human listening test and is close in quality to the target speech in the listening test. Audio samples can be found at https://home.ttic.edu/~jcchou/demo/avse/avse_demo.html.
format Preprint
id arxiv_https___arxiv_org_abs_2309_08030
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle AV2Wav: Diffusion-Based Re-synthesis from Continuous Self-supervised Features for Audio-Visual Speech Enhancement
Chou, Ju-Chieh
Chien, Chung-Ming
Livescu, Karen
Audio and Speech Processing
Computation and Language
Sound
Speech enhancement systems are typically trained using pairs of clean and noisy speech. In audio-visual speech enhancement (AVSE), there is not as much ground-truth clean data available; most audio-visual datasets are collected in real-world environments with background noise and reverberation, hampering the development of AVSE. In this work, we introduce AV2Wav, a resynthesis-based audio-visual speech enhancement approach that can generate clean speech despite the challenges of real-world training data. We obtain a subset of nearly clean speech from an audio-visual corpus using a neural quality estimator, and then train a diffusion model on this subset to generate waveforms conditioned on continuous speech representations from AV-HuBERT with noise-robust training. We use continuous rather than discrete representations to retain prosody and speaker information. With this vocoding task alone, the model can perform speech enhancement better than a masking-based baseline. We further fine-tune the diffusion model on clean/noisy utterance pairs to improve the performance. Our approach outperforms a masking-based baseline in terms of both automatic metrics and a human listening test and is close in quality to the target speech in the listening test. Audio samples can be found at https://home.ttic.edu/~jcchou/demo/avse/avse_demo.html.
title AV2Wav: Diffusion-Based Re-synthesis from Continuous Self-supervised Features for Audio-Visual Speech Enhancement
topic Audio and Speech Processing
Computation and Language
Sound
url https://arxiv.org/abs/2309.08030