Scribe v2

Speech recognition with word-level timestamps, diarization, entity detection, 90+ languages.

Scribe v2 is the batch transcription model: 90+ languages, word-level timestamps, diarization up to 32 speakers, entity detection, and keyterm prompting. I use it when the recording already exists.

Model spelling matches the official docs. For quotas, languages, and the latest changelog, check the official ElevenLabs website.