Plain-English Summary
ElevenLabs’ accurate batch transcription model for meetings, subtitles, and long recordings.
Scribe v2
ElevenLabs’ accurate batch transcription model for meetings, subtitles, and long recordings.
Scribe
Audio / Speech
GA
Open weights: No
API: Yes
Technical Notes
Speech recognition model supporting 90+ languages, long-form audio, speaker changes, pauses, and structured transcription.
Best For
Batch transcription, captions, subtitles, meeting and media archives
Watch Out For
Not optimized for live latency; billed separately from speech generation.
Model Specs
- Maker
- ElevenLabs
- Country
- USA
- Family
- Scribe
- Type
- Audio / Speech
- Release date
- 2026-01-09
- Date confidence
- Exact
- Status
- GA
- Context window
- Varies
- Max output
- Not listed
- Parameters
- Not disclosed
- Active parameters
- Not disclosed
- Architecture
- Proprietary automatic speech recognition model
- Reasoning
- Not applicable
- Tool calling
- No
- Structured output
- Yes — transcript metadata
- API available
- Yes
- Open weights
- No
- Self-hostable
- No
- License
- Proprietary
- Inputs
- Audio
- Outputs
- Text / timestamps / metadata
- Approx. price
- $0.22 per audio hour
- How to access
- Provider website or API
- Model / API ID
- scribe_v2
- Knowledge cutoff
- Not publicly disclosed
- Fine-tuning
- Key terms and transcription settings
- Completeness
- 81 %
- Verified
- Jul 18, 2026
- Snapshot date
- Jul 18, 2026
- Source
- Official source