ElevenLabs' most accurate speech-to-text model for recorded audio, with word-level timestamps and speaker diarization across 90+ languages. Transcribes via the /v1/audio/transcriptions endpoint.
View detailed pricing and capabilities for this provider.