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 4 · SOTA2 Research

Automatic Speech Recognition

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

ML-SUPERB 10-min (Normal)

17.1CER

26

Feb 26, 2026

1.13CER

26

May 11, 2026

10WER

26

Feb 26, 2026

LibriSpeech 100h clean (dev)

2.1WER

25

Jun 11, 2026

8.22MER

24

Jun 4, 2026

1.52WER

24

Jun 4, 2026

SPRING-INX Tamil

1D Component Count

24

May 28, 2026

CHIME-4 Real (test)

5.74WER

24

Feb 26, 2026

2.3WER

24

May 1, 2026

11.1WER

24

Feb 26, 2026

WenetSpeech (meeting)

4.63WER

23

May 28, 2026

9.25WER

23

Jun 11, 2026

4.82WER

23

Feb 26, 2026

7.76WER

22

Apr 28, 2026

Libri-Adapt Exp. 2

6.4WER (LIB Group 1)

22

Feb 26, 2026

TED-Lium 20 (test)

16.1WER

22

Feb 26, 2026

SPGISpeech 19 (test)

7.09WER

22

Feb 26, 2026

LibriSpeech clean 18

3.1WER

22

Feb 26, 2026

GigaSpeech (dev)

CR-CTC/AED, Zipformer-XL

0.099WER

22

Feb 26, 2026

0.514WER

22

Jun 9, 2026

manual audio seg

0.888LASER Score

21

Apr 9, 2026

0.831LASER

21

Apr 9, 2026

10.99WER/CER

21

Jun 11, 2026

LibriSpeech (LS) clean

Whisfusion + rerank

1.67WER

21

Jun 11, 2026

WenetSpeech Chuan hard

FireRedASR2S LLM

20.07CER

21

Apr 21, 2026

Page 4 of 38

Popular datasets

LibriSpeech FLEURS Common Voice AISHELL Multilingual ASR Corpus

Follow for updates

Join SOTA2 Contact