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 28 · SOTA2 Research

Automatic Speech Recognition

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

LibriSpeech clean + outdoor noise unseen (test)

9.06WER

5

Feb 26, 2026

News-Multilingual 15 languages (test)

12.8WER

5

Feb 26, 2026

Fleurs 30 languages (test)

8.16WER

5

Feb 26, 2026

Fleurs 20 languages (test)

6.62WER

5

Feb 26, 2026

Fleurs 12 languages (test)

4.9WER

5

Feb 26, 2026

MLC-SLM 11 languages (test)

12.74WER

5

Feb 26, 2026

CommonVoice 13 languages (test)

9.18WER

5

Jun 16, 2026

LibriSpeech ASR (test)

2WER

5

Mar 11, 2026

Multilingual LibriSpeech (test)

8.55WER

5

Feb 26, 2026

Spanish CallHome LDC96S35 (test)

17.95WER (%)

5

Feb 26, 2026

Multilingual LibriSpeech German (test)

3.93WER

5

Feb 26, 2026

LRS3 v0.4 (test)

Ours (filter-bank)

2.3WER

5

Feb 26, 2026

MCIF constrained, short audio

SeamlessM4T-v2-large (base)

21.49WER

4

Jul 8, 2026

6.5Error Rate (en)

4

Jul 7, 2026

LibriSpeech clean other EuroParlST CoVoST2

SpeechMapper + LoRA (C)

10.5MCIF

4

Jul 3, 2026

MCIF short-form

88.77ASR Accuracy

4

Jun 26, 2026

FLEURS South-Asia (14 languages from the 8th Schedule) (test)

0.534WER

4

Jun 24, 2026

MUSAN additive noise on LibriSpeech (dev-other)

Greedy decoding

6.62Greedy Score

4

Jun 23, 2026

CHiME estimated diarization 8 (eval)

36.81tcpWER (CHiME-6)

4

Jun 12, 2026

OGI Spontaneous (test)

11.06WER (Overall)

4

Jun 11, 2026

48.09WER (Very Low)

4

Jun 11, 2026

7WER

4

May 26, 2026

6.9WER

4

May 19, 2026

8.7WER

4

May 19, 2026

5.8WER

4

May 19, 2026

Page 28 of 38

Popular datasets

LibriSpeech FLEURS Common Voice AISHELL Multilingual ASR Corpus

Follow for updates

Join SOTA2 Contact