ModelRadar
/

transcribe-03

Cohere🇨🇦 CAMajor lab·

Audioopen weights

No description yet — the model card or announcement hasn't been summarised.

Modalities
text → text

1 variant

2026HF

Sources

HuggingFace

Benchmarks

BenchmarkDomainValueEvidence
Earnings-22Open ASR Leaderboard (Langform)audio-stt12.66 % WER68
Open ASR LeaderboardOpen ASR Leaderboard (Langform)audio-stt9.733125 % WER58
Common Voice (ASR)Open ASR Leaderboardaudio-stt5.28 % WER52
Common Voice (ASR)Open ASR Leaderboardaudio-stt2.87 % WER52
Common Voice (ASR)Open ASR Leaderboardaudio-stt2.79 % WER52
Common Voice (ASR)Open ASR Leaderboardaudio-stt2.51 % WER52
FLEURSOpen ASR Leaderboardaudio-stt4.33 % WER52
FLEURSOpen ASR Leaderboardaudio-stt4.24 % WER52
FLEURSOpen ASR Leaderboardaudio-stt3.33 % WER52
FLEURSOpen ASR Leaderboardaudio-stt3.26 % WER52
FLEURSOpen ASR Leaderboardaudio-stt2.33 % WER52
TED-LIUM 3Open ASR Leaderboard (Langform)audio-stt2.23 % WER49

Vendor-reported numbers are marked and count with factor 0.7 in rankings. The evidence score says how much a benchmark still tells you today.

Provenance

Detected
Aug 25, 2026
First source
openasr
Description
not yet
Editorially reviewed
—