IBMFermion ResearchAudio matchupSpeech to text
Granite Speech 5.0 470M TurboCTC NCvs.Phonon-2
Which model is better?
Granite Speech 5.0 470M TurboCTC NC is slightly ahead in independent tests, but the lead is not certain.
Pick Granite Speech 5.0 470M TurboCTC NC for
Pick Phonon-2 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
Granite Speech 5.0 470M TurboCTC NC slightly aheadGranite Speech 5.0 470M TurboCTC NC is clearly ahead in 0 of 10 subtests and slightly in 7, Phonon-2 in 0 clearly and slightly in 2. 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: URGENT24
lower is better, as of October 28, 2026
Open ASR English short-form WER: Private (scripted)
lower is better, as of October 28, 2026
Show all 10 subtests
| Subtest | Granite Speech 5.0 470M TurboCTC NC | Phonon-2 | Verdict |
|---|---|---|---|
| Open ASR English short-form WER: Voxpopuli-AA-Cleaned (lower is better) | 4.7 % | 2.5 % | Phonon-2 slightly ahead |
| Open ASR English short-form WER: URGENT24 (lower is better) | 10.2 % | 14.9 % | Granite Speech 5.0 470M TurboCTC NC slightly ahead |
| Open ASR English short-form WER: Private (scripted) (lower is better) | 3.8 % | 5.2 % | Granite Speech 5.0 470M TurboCTC NC slightly ahead |
| Open ASR English short-form WER: AMI-Cleaned (lower is better) | 7.3 % | 9.3 % | Granite Speech 5.0 470M TurboCTC NC slightly ahead |
| Open ASR English short-form WER: LS Clean (lower is better) | 1.5 % | 1.7 % | Granite Speech 5.0 470M TurboCTC NC slightly ahead |
| Open ASR English short-form WER: Earnings22-Cleaned-AA-chunked (lower is better) | 8.2 % | 7.0 % | Phonon-2 slightly ahead |
| Open ASR English short-form WER: Voice Arena Monsoon (lower is better) | 4.3 % | 4.9 % | Granite Speech 5.0 470M TurboCTC NC slightly ahead |
| Open ASR English short-form WER: Average WER (11-set macro-average) (lower is better) | 6.0 % | 6.9 % | Granite Speech 5.0 470M TurboCTC NC slightly ahead |
| Open ASR English short-form WER: Private (conversational) (lower is better) | 13.5 % | 14.1 % | Granite Speech 5.0 470M TurboCTC NC slightly ahead |
| Open ASR English short-form WER: Gigaspeech-Cleaned (lower is better) | 8.4 % | 8.4 % | 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 NC | Phonon-2 | |
|---|---|---|
| Task | Speech to text | Speech to text |
| Developer | IBM | Fermion Research |
| Status | Open weights | Open weights |
| Released | August 25, 2026 | n/a |
| Weights | open | open |
| Runs locally | yes | yes |
| Parameters | 0.47 B | n/a |
| License | CC-BY-NC-SA-4.0 | CC-BY-4.0 |
| Popularity (out of 100) | 50 | 44 |