Scribe V3
Transcribe Multilingual Audio With ElevenLabs Scribe v3
Turn recorded interviews or meetings into speaker-diarised transcripts across 102 languages using ElevenLabs Scribe v3.
- Beginner
- 3 steps
- Paid (free tier available)
By TWK Guides DeskUpdated 18 Nov 2026
Every step checked against ElevenLabs's own docs on 18 Nov 2026
Overview
ElevenLabs Scribe v3 cuts word errors by 27% over Scribe v2 and supports both recorded file transcription and real-time Scribe v3 Live streams across 102 languages.
Before you start
- An ElevenLabs account with Speech to Text access
- An audio or video file (MP3, WAV, M4A, or MP4) with clear speech
Step by step
0 of 3 done
Your ticks stay in this browser.
Step 1 of 3
Open Speech to Text in the ElevenLabs workspace
Start in the Speech to Text tab so you can choose between recorded batch processing and live streaming.
Do this
0 of 2 done
What you should see
The model selector shows Scribe v3 active for your workspace.
If it didn't work
If Scribe v3 does not appear, refresh your workspace permissions or check your plan tier.
Checked againstelevenlabs.io
Frequently asked questions
When should I use Scribe v3 Live instead of Scribe v3?
Use Scribe v3 for recorded files where maximum accuracy matters, and Scribe v3 Live when you need real-time captions or voice agent input.
