Transcribe audio to text using Scribe.

POST multipart form data with audio file. Returns transcription text.

{
  "body": {
    "audio": "file (required, audio file)"
  }
}

Get a transcript by ID.

Replace :transcriptionId. Returns the stored transcript.

{
  "path": {
    "transcriptionId": "string"
  }
}

Convert text to speech audio with a specific voice.

Replace :voiceId with a voice ID. POST JSON body with text. Returns audio data.

{
  "path": {
    "voiceId": "string (voice ID, use GET /v1/voices to list available)"
  },
  "body": {
    "text": "string (required, text to speak)",
    "model_id": "string (optional, e.g. 'eleven_multilingual_v2')"
  }
}

Stream text-to-speech audio.

Like /v1/text-to-speech/:voiceId but streams the audio response.

{
  "path": {
    "voiceId": "string"
  },
  "body": {
    "text": "string (required)",
    "model_id": "string (optional)"
  }
}

List available text-to-speech voices.

GET with no params. Returns list of available voices with IDs and metadata.