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

Audio-Visual Speech Recognition research benchmarks · SOTA2 Research

Audio-Visual Speech Recognition

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

0.68WER

77

Apr 2, 2026

LRS3 clean (test)

0.72WER

77

Mar 5, 2026

LRS-3 Babble noise at 0dB SNR (test)

1.9WER

53

Jun 5, 2026

1.3WER

34

Feb 26, 2026

LRS3 30h labeled low-resource (test)

1.8WER

22

Mar 4, 2026

LRS3 Clean original (test)

0.65WER

21

Jun 5, 2026

LRS3 Noisy w.r.t Missing Modality 0.3 Masking Ratio (test)

5.77WER (%)

16

Jun 5, 2026

LRS3 Noisy w.r.t Missing Modality 0.1 Masking Ratio (test)

4.07Word Error Rate (%)

16

Jun 5, 2026

LRS3 Noisy w.r.t Viewpoint 15° Offset (test)

4.05WER

16

Jun 5, 2026

LRS3 Noisy w.r.t Viewpoint 5° Offset (test)

3.86WER

16

Jun 5, 2026

2.2WER

16

Apr 2, 2026

0.008WER

14

Mar 24, 2026

MISP 2021 (test)

0.188CER

13

Jun 5, 2026

0.385WER

12

Feb 26, 2026

GER w/ Auto-AVSR

43Overall Score

10

Feb 27, 2026

LRS2 50% visual occlusion (test)

ASR + VSR oracle onb / ocp

6.4WER (Overall)

10

Feb 27, 2026

0.009WER

10

Feb 26, 2026

MuAViC Noise environment (test)

49.5Accuracy (En)

9

Feb 27, 2026

MuAViC Clean environment (test)

2.5En Acc

9

Feb 27, 2026

Simulated LRS2 19 (test)

5.2WER (-5 dB)

8

Jul 9, 2026

AV-HuBERT + CMA + MoHAVE

4.2Average Error Rate

8

Feb 27, 2026

LRS3 433 h 0 dB SNR

1.7WER

7

Mar 5, 2026

LRS3 433 h 5 dB SNR

1.3WER

7

Mar 5, 2026

LRS3 Pixelated face

26WER (Babble, -10 dB)

7

Feb 27, 2026

LRS3 Occlusion by hands

26.6WER (Babble, -10 dB)

7

Feb 27, 2026

Page 1 of 3

PreviousNext

Popular datasets

LRS3 LRS2 LRS-3 MuAViC FLEURS

Follow for updates

Join SOTA2 Contact