$schema: https://json-schema.org/draft/2020-12/schema
$id: https://raw.githubusercontent.com/api-evangelist/google-cloud-speech-to-text/refs/heads/main/json-schema/json-schema.yml
title: Google Cloud Speech-to-Text Recognition Request
description: A speech recognition request for the Google Cloud Speech-to-Text API
type: object
properties:
  config:
    type: object
    description: Configuration for the recognition request
    properties:
      encoding:
        type: string
        enum:
          - LINEAR16
          - FLAC
          - MULAW
          - AMR
          - AMR_WB
          - OGG_OPUS
          - SPEEX_WITH_HEADER_BYTE
          - WEBM_OPUS
          - MP3
        description: Encoding of the supplied audio data
      sampleRateHertz:
        type: integer
        minimum: 8000
        maximum: 48000
        description: Sample rate in Hertz of the audio data
      audioChannelCount:
        type: integer
        minimum: 1
        description: Number of channels in the input audio
      languageCode:
        type: string
        pattern: ^[a-z]{2}(-[A-Z]{2})?$
        description: Language of the supplied audio in BCP-47 format
      alternativeLanguageCodes:
        type: array
        items:
          type: string
        description: Additional languages that may be present in the audio
      maxAlternatives:
        type: integer
        minimum: 1
        maximum: 30
        description: Maximum number of recognition hypotheses to return
      enableWordTimeOffsets:
        type: boolean
        description: If true, word-level time offsets are included in the response
      enableWordConfidence:
        type: boolean
        description: If true, word-level confidence is included in the response
      enableAutomaticPunctuation:
        type: boolean
        description: If true, automatic punctuation is added to the transcription
      enableSpokenPunctuation:
        type: boolean
        description: If true, spoken punctuation is detected
      model:
        type: string
        enum:
          - default
          - command_and_search
          - phone_call
          - video
          - latest_long
          - latest_short
          - medical_dictation
          - medical_conversation
        description: Speech recognition model to use
      useEnhanced:
        type: boolean
        description: If true, use an enhanced model for recognition
    required:
      - languageCode
  audio:
    type: object
    description: The audio data to be recognized
    properties:
      content:
        type: string
        contentEncoding: base64
        description: Base64-encoded audio data
      uri:
        type: string
        format: uri
        description: URI that points to the audio file (GCS URI)
    description: Audio data for recognition
required:
  - config
  - audio