List your company

Automatic Speech Recognition on LRS3 Clean original (test)

0.68WERWhisper-Large, Fine-tuned w/o noise
0.59921.14461.692.2354Jun 14, 2024

Evaluation results

MethodLinks
Whisper-Large, Fine-tuned w/o noiseBeam search=1, Labeled AV Training Hrs=1,7592024.060.68
Fast ConformerNoise Dataset=NoiseX, Unlabeled AV Training Hrs=3,1162024.060.7
Llama-AVSRNoise Dataset=NoiseX, Unlabeled AV Training Hrs=1,759, Labeled AV Training Hrs=1,7592024.060.81
Whisper-Large, Fine-tuned w/ noiseBeam search=1, Labeled AV Training Hrs=1,7592024.060.85
LP ConformerNoise Dataset=NoiseX, Labeled AV Training Hrs=100k2024.060.9
AutoAVSRNoise Dataset=NoiseX, Unlabeled AV Training Hrs=1,9022024.061
AutoAVSRNoise Dataset=NoiseX, Unlabeled AV Training Hrs=3,4482024.061
Whisper-Large, Fine-tuned w/o noiseBeam search=1, Labeled AV Training Hrs=4332024.061
Llama-AVSRNoise Dataset=NoiseX, Unlabeled AV Training Hrs=1,759, Labeled AV Training Hrs=4332024.061.1
Whisper-Large, Fine-tuned w/ noiseBeam search=1, Labeled AV Training Hrs=4332024.061.1
Whisper-Large, Fine-tuned w/o noiseBeam search=15, Labeled AV Training Hrs=4332024.061.3
u-HuBERTNoise Dataset=LRS3+, Unlabeled AV Training Hrs=1,759, Labeled AV Training Hrs=4332024.061.4
Whisper-Large, Fine-tuned w/o noiseBeam search=15, Labeled AV Training Hrs=1,7592024.061.5
VIT 3DNoise Dataset=NoiseX, Labeled AV Training Hrs=90k2024.061.6
Fast ConformerNoise Dataset=NoiseX, Unlabeled AV Training Hrs=4352024.061.6
AV-HUBERTNoise Dataset=LRS3+, Unlabeled AV Training Hrs=1,759, Labeled AV Training Hrs=4332024.061.6
Whisper-Large, Fine-tuned w/ noiseBeam search=15, Labeled AV Training Hrs=1,7592024.062
Whisper-Large Zero-shotBeam search=152024.062.1
Whisper-Large Zero-shotBeam search=12024.062.3
Whisper-Large, Fine-tuned w/ noiseBeam search=15, Labeled AV Training Hrs=4332024.062.3
AV2vecNoise Dataset=MUSAN, Unlabeled AV Training Hrs=433, Labeled AV Training Hrs=4332024.062.7