Lipreading On Lrs2

Word Error Rate (WER)

Results

Performance results of various models on this benchmark

Comparison Table

Model Name	Word Error Rate (WER)
es3-evolving-self-supervised-learning-of	26.7
es3-evolving-self-supervised-learning-of	31.4
asr-is-all-you-need-cross-modal-distillation	53.2
unified-speech-recognition-a-single-model-for	15.4
sub-word-level-lip-reading-with-visual	22.6
audio-visual-recognition-of-overlapped-speech	48.86
visual-speech-recognition-for-multiple	25.5
distinguishing-homophenes-using-multi-head-1	44.5
es3-evolving-self-supervised-learning-of	24.6
syncvsr-data-efficient-visual-speech	28.9
leveraging-uni-modal-self-supervised-learning-1	43.2
es3-evolving-self-supervised-learning-of	28.7
deep-audio-visual-speech-recognition	48.3
end-to-end-audio-visual-speech-recognition	39.1
spatio-temporal-fusion-based-convolutional	51.7
auto-avsr-audio-visual-speech-recognition	14.6
deep-audio-visual-speech-recognition	54.7
syncvsr-data-efficient-visual-speech	16.5
jointly-learning-visual-and-auditory-speech	18.6
es3-evolving-self-supervised-learning-of	29.3
hearing-lips-improving-lip-reading-by	65.29
audio-visual-speech-recognition-with-a-hybrid	50
es3-evolving-self-supervised-learning-of	30.7
sub-word-level-lip-reading-with-visual	28.9
visual-speech-recognition-for-multiple	32.9