audio.transcribe.submit
Submit audio files for transcription to convert speech to text. Supports multiple formats, auto-detects languages, and optionally identifies speakers.
Instructions
Submit an audio file URL for speech-to-text transcription. Returns a transcript_id to check status and retrieve results. Supports MP3, WAV, M4A, FLAC, OGG, WebM. 99 languages auto-detected. Optional speaker diarization (AssemblyAI)
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| audio_url | Yes | Publicly accessible URL of the audio file to transcribe (MP3, WAV, M4A, FLAC, OGG, WebM) | |
| model | No | Speech model: "universal-2" (default, fast, 99 languages) or "universal-3-pro" (highest accuracy, promptable) | |
| language_code | No | Language code (e.g. "en", "es", "de", "fr", "ja"). Auto-detected if omitted | |
| speaker_labels | No | Enable speaker diarization — detect who said what (default false) |