{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "#/components/schemas/TranscriptionResponse",
  "title": "TranscriptionResponse",
  "type": "object",
  "properties": {
    "text": {
      "type": "string",
      "description": "The transcribed text"
    },
    "task": {
      "type": "string",
      "description": "The task performed (transcribe)"
    },
    "language": {
      "type": "string",
      "description": "The detected or specified language"
    },
    "duration": {
      "type": "number",
      "description": "The duration of the audio in seconds"
    },
    "words": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "word": {
            "type": "string"
          },
          "start": {
            "type": "number"
          },
          "end": {
            "type": "number"
          }
        }
      },
      "description": "Word-level timestamps (when requested)"
    },
    "segments": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "id": {
            "type": "integer"
          },
          "start": {
            "type": "number"
          },
          "end": {
            "type": "number"
          },
          "text": {
            "type": "string"
          }
        }
      },
      "description": "Segment-level timestamps (when requested)"
    }
  }
}