Scribe v3 vs Scribe v3 Live
Compare documented capabilities and access. A dash means a comparable figure is unavailable.
| Specification | ||
|---|---|---|
| Task | Speech recognition | Speech recognition |
| Published capabilities | Transcribes recorded audio. ElevenLabs reports 27% fewer word errors than Scribe v2 across 102 languages in its launch evaluation. | Transcribes live audio with partial results and turn prediction. ElevenLabs reports 43% fewer errors than Scribe v2 Realtime in its launch evaluation. |
| Access | ElevenLabs Speech to Text API | ElevenLabs Speech to Text API |
| Result | Transcript with speaker labels | Streaming partial and committed transcripts |
| Sources | Publisher documentation · Additional documentation | Publisher documentation · Additional documentation |
| Inputs | Audio | Audio |
| Weights | Weights unconfirmed | Weights unconfirmed |
| Released | Nov 9, 2026 | Nov 9, 2026 |