MODELS
by ElevenLabs
Speech-to-text model for transcription across 90+ languages with word-level timestamps, speaker diarization, keyterm prompting, and audio tagging.
ElevenLabs Scribe v2 is a speech-to-text model from ElevenLabs. The supplied ElevenLabs documentation and help sources describe it as supporting transcription across 90+ languages, with word-level timestamps, speaker diarization, keyterm prompting, and audio tagging. It is documented in ElevenLabs model documentation and Speech-to-Text help materials.
Use when you need speech-to-text transcription across 90+ languages. Use when transcription output needs word-level timestamps speaker diarization keyterm prompting or audio tagging.
Last updated Jun 9, 2026
