SpeechToTextRequest

{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "#/components/schemas/SpeechToTextRequest",
  "title": "SpeechToTextRequest",
  "type": "object",
  "required": [
    "file"
  ],
  "properties": {
    "file": {
      "type": "string",
      "format": "binary",
      "description": "The audio file to transcribe. Supports formats including MP3, WAV, FLAC, OGG, and M4A."
    },
    "model_id": {
      "type": "string",
      "description": "The identifier of the speech-to-text model to use for transcription."
    },
    "language_code": {
      "type": "string",
      "description": "Language code in ISO 639-1 format to hint the expected language of the audio content."
    },
    "tag_audio_events": {
      "type": "boolean",
      "description": "Whether to tag non-speech audio events such as music, laughter, or applause in the transcription output.",
      "default": false
    },
    "timestamps_granularity": {
      "type": "string",
      "description": "The level of timestamp granularity to include in the response.",
      "enum": [
        "none",
        "word",
        "character"
      ]
    }
  }
}