Model profile

Scribe V2

Speech to Text

ElevenLabs speech-to-text for transcripts, subtitles, captions, and long-form audio cleanup.

First announced

March 3, 2025

Available apps

1 product pages

Model background

Documented by ElevenLabs in its March 3, 2025 changelog when Scribe v2 became available, then expanded in the March 6, 2026 product announcement for batch transcription, subtitles, captions, and multilingual speech-to-text workflows.

Lab: ElevenLabs

First announced: March 3, 2025

Best uses

  • Batch transcription for long-form audio
  • Useful for subtitles, captions, and searchable transcripts
  • Optional diarization and audio-event tagging for production cleanup

Output types

TextWord timestamps

Primary sources

Available in creator apps

Speech to Text

Turn uploaded audio into transcripts and caption-ready text.