Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

AVSpeech dataset and benchmarks · SOTA2 Research

Research Datasets

AVSpeech

Benchmarks

Task NameDataset NameSOTA MethodSort

Task NameDataset NameSOTA ResultTrendResults

Speech-to-Portrait

AVSpeech (test)

31.26L1 Error

6

Visual Acoustic Matching

AVSpeech-Rooms unseen environments (test)

0.071RTE (s)

5

Audio-Visual Speech Recognition

AVSpeech (1,000 manually filtered samples)

25WER

4

Audio-Visual Speech Extraction

AVSpeech

10.2SI-SDR (dB)

1

Popular tasks

Audio-Visual Speech Recognition Visual Acoustic Matching Speech-to-Portrait Audio-Visual Speech Extraction

Follow for updates

Join SOTA2 Contact