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

Automatic Speech Recognition research benchmarks – Page 3 · SOTA2 Research

Automatic Speech Recognition

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

93.7WAR

34

Apr 10, 2026

LibriSpeech Clean other (test)

5.08WER

34

Apr 21, 2026

0.83WER

34

May 25, 2026

ASRU 2019 (test)

Interactive ASR

2.88Match Error Rate (MER)

32

Apr 13, 2026

H-O out-of-domain Sudanese dataset (holdout)

51.6WER

32

Feb 26, 2026

57.1WER

32

Feb 26, 2026

UIT-ViMD (test)

12.58WER

31

May 28, 2026

Context biasing + text replacement oracle

5.9WER

31

Mar 5, 2026

0.6WER

31

May 1, 2026

0.175Average Error Rate

30

Jul 7, 2026

Voicebank+Demand

1.71WER

30

Feb 26, 2026

13.5WER

30

Feb 26, 2026

0.133WER

30

Feb 26, 2026

Audio Flamingo 3 + CoAT

1.84WER

30

Jun 18, 2026

TED-LIUM (test)

2.9WER

30

Mar 9, 2026

2.27CER

29

Mar 24, 2026

LibriSpeech other Speech Noise - Reverb (test)

30.1WER

28

Mar 16, 2026

LibriSpeech clean Speech Noise - Reverb (test)

19.8WER

28

Mar 16, 2026

LibriSpeech other Speech Noise - Additive (test)

13.1WER

28

Mar 16, 2026

LibriSpeech clean Speech Noise - Additive (test)

6.3WER

28

Mar 16, 2026

CHiME-4 (dev-simu)

5.6WER (%)

28

Jun 5, 2026

26WER (%)

28

Apr 27, 2026

3.5CER (DS2)

27

Feb 26, 2026

NIST CTS CallHome 2000

2016 Microsoft system

11WER

27

Feb 26, 2026

LRS3 30h labeled low-resource (test)

1.5WER

26

Mar 4, 2026

Page 3 of 38

Popular datasets

LibriSpeech FLEURS Common Voice AISHELL Multilingual ASR Corpus

Follow for updates

Join SOTA2 Contact