IBMMeta Superintelligence LabsAudio matchupSpeech to text
Granite Speech 5.0 470M TurboCTCvs.Muse Voice Transcribe
Which model is better?
Granite Speech 5.0 470M TurboCTC and Muse Voice Transcribe each lead in different tests.
Pick Granite Speech 5.0 470M TurboCTC for
Pick Muse Voice Transcribe for
Shared tests
Independent evaluators and leaderboards only, the newest shared snapshot per evaluator. With 95 % ranges a lead counts when they do not overlap; for hit rates, when it holds in a 95 % test. Without any uncertainty figure at most “slightly ahead”.
Hugging Face Open ASR
depends on the subtestMuse Voice Transcribe is clearly ahead in 0 of 9 subtests and slightly in 5, Granite Speech 5.0 470M TurboCTC in 0 clearly and slightly in 3. 1 is level.
Open ASR English short-form WER: Voxpopuli-AA-Cleaned
lower is better, as of October 28, 2026
Open ASR English short-form WER: AMI-Cleaned
lower is better, as of October 28, 2026
Open ASR English short-form WER: Voice Arena Monsoon
lower is better, as of October 28, 2026
Show all 9 subtests
| Subtest | Granite Speech 5.0 470M TurboCTC | Muse Voice Transcribe | Verdict |
|---|---|---|---|
| Open ASR English short-form WER: Voxpopuli-AA-Cleaned (lower is better) | 4.9 % | 2.0 % | Muse Voice Transcribe slightly ahead |
| Open ASR English short-form WER: AMI-Cleaned (lower is better) | 7.7 % | 9.2 % | Granite Speech 5.0 470M TurboCTC slightly ahead |
| Open ASR English short-form WER: Voice Arena Monsoon (lower is better) | 4.8 % | 4.2 % | Muse Voice Transcribe slightly ahead |
| Open ASR English short-form WER: Gigaspeech-Cleaned (lower is better) | 8.7 % | 7.8 % | Muse Voice Transcribe slightly ahead |
| Open ASR English short-form WER: Earnings22-Cleaned-AA-chunked (lower is better) | 6.8 % | 6.0 % | Muse Voice Transcribe slightly ahead |
| Open ASR English short-form WER: URGENT24 (lower is better) | 11.8 % | 13.0 % | Granite Speech 5.0 470M TurboCTC slightly ahead |
| Open ASR English short-form WER: Private (scripted) (lower is better) | 4.1 % | 4.5 % | Granite Speech 5.0 470M TurboCTC slightly ahead |
| Open ASR English short-form WER: Private (conversational) (lower is better) | 14.0 % | 13.0 % | Muse Voice Transcribe slightly ahead |
| Open ASR English short-form WER: Average WER (11-set macro-average) (lower is better) | 6.4 % | 6.3 % | level |
Source · latest October 28, 2026
All data side by side
Prices are verbatim from the provider, because audio models bill per minute, character or token.
| Granite Speech 5.0 470M TurboCTC | Muse Voice Transcribe | |
|---|---|---|
| Task | Speech to text | Speech to text |
| Developer | IBM | Meta Superintelligence Labs |
| Status | Open weights | Active |
| Released | August 25, 2026 | September 1, 2026 |
| Weights | open | n/a |
| Runs locally | yes | no |
| Parameters | 0.47 B | n/a |
| License | Apache-2.0 | n/a |
| Price | n/a | $0.18 per audio hour ($3 per 1,000 audio minutes) |
| Popularity (out of 100) | 52 | 0 |