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 27 · SOTA2 Research

Automatic Speech Recognition

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

OpenSLR Bengali ASR

Multi-Level Embedding Conformer

10.01WER (%)

5

Feb 26, 2026

AIShell2 (test)

Step-Audio 2 mini

2.36WER

5

Feb 26, 2026

In-house EN music (test)

24.61WER

5

Feb 26, 2026

In-house EN speech (test)

9.42WER

5

Feb 26, 2026

In-house ZH domain H (test)

3.59WER

5

Feb 26, 2026

In-house ZH domain G (test)

4.2WER

5

Feb 26, 2026

In-house ZH domain F (test)

6.25WER

5

Feb 26, 2026

In-house ZH domain E (test)

6.57WER

5

Feb 26, 2026

In-house ZH domain D (test)

9.33WER

5

Feb 26, 2026

In-house ZH domain C (test)

4.31WER

5

Feb 26, 2026

In-house ZH domain B (test)

5.38WER

5

Feb 26, 2026

In-house ZH domain A (test)

9.9WER

5

Feb 26, 2026

S3 Tokenizer(FSQ)

7.29WER

5

Feb 26, 2026

SlideSpeech target

Adapted model (audio)

10.37WER

5

Feb 26, 2026

DefinedAI Insurance target domain (test)

Adapted model (audio)

7.92WER

5

Feb 26, 2026

DefinedAI Banking target domain (test)

Adapted model (audio)

9.92WER

5

Feb 26, 2026

CHiME-4 single-channel (et_real)

WavLM (Weighted-sum) + Joint FT. WavLM

3.5WER

5

Feb 26, 2026

CHiME-4 single-channel (dt_real)

2WER

5

Feb 26, 2026

CHiME-4 single-channel (dt_simu)

WavLM (Weighted-sum) + Joint FT. WavLM

3.1WER

5

Feb 26, 2026

ProtoDisent-TTS

2.32WER

5

Feb 26, 2026

TORGO (Mod-Severe)

ProtoDisent-TTS

0.117WER

5

Feb 26, 2026

ProtoDisent-TTS

22.15WER

5

Feb 26, 2026

FLEURS SSA 102 (test)

9.1AFR CER

5

Feb 26, 2026

MLC-SLM CV (test)

11.47WER/CER

5

Feb 26, 2026

LibriSpeech clean + indoor noise seen noise (test)

8.86WER

5

Feb 26, 2026

Page 27 of 38

Popular datasets

LibriSpeech FLEURS Common Voice AISHELL Multilingual ASR Corpus

Follow for updates

Join SOTA2 Contact