# OpenAI Integration Audio Schemas (Speech and Transcription)

# Speech (TTS) Request
OpenAISpeechRequest:
  type: object
  required:
    - model
    - input
  properties:
    model:
      type: string
      description: Model identifier (e.g., tts-1, tts-1-hd)
      example: tts-1
    input:
      type: string
      description: Text to convert to speech
    voice:
      type: string
      description: Voice to use
      enum: [alloy, echo, fable, onyx, nova, shimmer]
    response_format:
      type: string
      enum: [mp3, opus, aac, flac, wav, pcm]
    speed:
      type: number
      minimum: 0.25
      maximum: 4.0
    stream_format:
      type: string
      enum: [sse]
      description: Set to 'sse' for streaming
    # Bifrost-specific
    fallbacks:
      type: array
      items:
        type: string

# Transcription Request
OpenAITranscriptionRequest:
  type: object
  required:
    - model
    - file
  properties:
    model:
      type: string
      description: Model identifier (e.g., whisper-1)
      example: whisper-1
    file:
      type: string
      format: binary
      description: Audio file to transcribe
    language:
      type: string
      description: Language of the audio (ISO 639-1)
    prompt:
      type: string
      description: Prompt to guide transcription
    response_format:
      type: string
      enum: [json, text, srt, verbose_json, vtt]
    temperature:
      type: number
      minimum: 0
      maximum: 1
    timestamp_granularities:
      type: array
      items:
        type: string
        enum: [word, segment]
    stream:
      type: boolean
    # Bifrost-specific
    fallbacks:
      type: array
      items:
        type: string

# Responses reuse inference schemas
OpenAISpeechResponse:
  $ref: '../../inference/speech.yaml#/SpeechResponse'

OpenAISpeechStreamResponse:
  $ref: '../../inference/speech.yaml#/SpeechStreamResponse'

OpenAITranscriptionResponse:
  $ref: '../../inference/transcription.yaml#/TranscriptionResponse'

OpenAITranscriptionStreamResponse:
  $ref: '../../inference/transcription.yaml#/TranscriptionStreamResponse'