Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

LRS3-TED

Benchmarks

Task NameDataset NameSOTA ResultTrend
Automatic Speech RecognitionLRS3-TED (test)
WER7.2
13
Automatic Speech RecognitionLRS3-TED Noisy (test)
WER27.7
9
Video-to-speech synthesisLRS3-TED (test)
UTMOS4.031
7
Visual Speech RecognitionLRS3-TED Full (test)
WER55.1
2
Showing 4 of 4 rows