Papers › ReVISE: Self-Supervised Speech Resynthesis with Visual Input for Universal and...

ReVISE: Self-Supervised Speech Resynthesis with Visual Input for Universal and Generalized Speech Enhancement

21 Dec 2022arXiv:2212.11377archive 2025-07-28

Wei-Ning Hsu, Tal Remez, Bowen Shi, Jacob Donley, Yossi Adi

Prior works on improving speech quality with visual input typically study each type of auditory distortion separately (e.g., separation, inpainting, video-to-speech) and present tailored algorithms. This paper proposes to unify these subjects and study Generalized Speech Enhancement, where the goal is not to reconstruct the exact reference clean signal, but to focus on improving certain aspects of speech. In particular, this paper concerns intelligibility, quality, and video synchronization. We cast the problem as audio-visual speech resynthesis, which is composed of two steps: pseudo audio-visual speech recognition (P-AVSR) and pseudo text-to-speech synthesis (P-TTS). P-AVSR and P-TTS are connected by discrete units derived from a self-supervised speech model. Moreover, we utilize self-supervised audio-visual speech model to initialize P-AVSR. The proposed model is coined ReVISE. ReVISE is the first high-quality model for in-the-wild video-to-speech synthesis and achieves superior performance on all LRS3 audio-visual enhancement tasks with a single model. To demonstrates its applicability in the real world, ReVISE is also evaluated on EasyCom, an audio-visual benchmark collected under challenging acoustic conditions with only 1.6 hours of training data. Similarly, ReVISE greatly suppresses noise and improves quality. Project page: https://wnhsu.github.io/ReVISE.

PaperPDF

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Audio-Visual Speech RecognitionResynthesisSpeech EnhancementSpeech RecognitionSpeech SynthesisText to SpeechText-To-Speech SynthesisVideo SynchronizationVisual Speech Recognitionspeech-recognitiontext-to-speech

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Speech Enhancement EasyCom ReVISE (ch2) Audio Quality MOS 4.19 #2 of 6 Archive leaderboard report
Speech Enhancement EasyCom ReVISE (bf) Audio Quality MOS 4.11 #3 of 6 Archive leaderboard report
Speech Enhancement EasyCom Demucs (ch2) Audio Quality MOS 2.95 #4 of 6 Archive leaderboard report
Speech Enhancement EasyCom Demucs (bf) Audio Quality MOS 2.39 #5 of 6 Archive leaderboard report
Speech Recognition EasyCom ReVISE (bf) WER (%) 52.1 #1 of 5 Archive leaderboard report
Speech Recognition EasyCom ReVISE (ch2) WER (%) 55.0 #2 of 5 Archive leaderboard report
Speech Recognition EasyCom Demucs (bf) WER (%) 69.8 #4 of 5 Archive leaderboard report
Speech Recognition EasyCom Demucs (ch2) WER (%) 86.8 #5 of 5 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections