Google Speech Generation API

openapi: 3.1.0
info:
  title: Google Books About Speech Generation API
  description: API for retrieving Bookshelf and Volume resources from Google Books
  version: 1.0.0
  contact:
    name: Google Books API
    url: https://developers.google.com/books
servers:
- url: https://www.googleapis.com/books/v1
  description: Google Books API v1
tags:
- name: Speech Generation
  description: Operations related to Speech Generation
paths:
  /models/{ttsModel}:generateContent:
    post:
      tags:
      - Speech Generation
      summary: Google Generate Speech (Text-to-Speech)
      description: "Transform text input into single-speaker or multi-speaker audio using native text-to-speech (TTS) \ngeneration capabilities. TTS is controllable through natural language to guide style, accent, pace, \nand tone of the audio.\n\n**Capabilities:**\n- Single-speaker or multi-speaker audio (up to 2 speakers)\n- 30 voice options with different characteristics (bright, upbeat, informative, etc.)\n- 24 supported languages with automatic language detection\n- Controllable style, tone, accent, and pace via prompts\n- Audio output in PCM format (24kHz, 16-bit, mono)\n\n**Note**: TTS models accept text-only inputs and produce audio-only outputs.\n"
      operationId: generateSpeech
      parameters:
      - name: ttsModel
        in: path
        required: true
        description: The TTS model to use for speech generation
        schema:
          type: string
          enum:
          - gemini-2.5-flash-preview-tts
          - gemini-2.5-pro-preview-tts
          example: gemini-2.5-flash-preview-tts
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/SpeechGenerationRequest'
      responses:
        '200':
          description: Successful response with generated audio
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/SpeechGenerationResponse'
        '400':
          description: Bad request
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
        '401':
          description: Unauthorized - Invalid API key
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
        '429':
          description: Too many requests
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
      x-microcks-operation:
        delay: 100
        dispatcher: FALLBACK
        dispatcherRules: SuccessExample
components:
  schemas:
    Content:
      type: object
      properties:
        role:
          type: string
          enum:
          - user
          - model
          description: The role of the content producer (optional for single-turn, required for multi-turn)
        parts:
          type: array
          description: Ordered parts that constitute a single message
          items:
            $ref: '#/components/schemas/Part'
      required:
      - parts
    SpeechConfig:
      type: object
      description: Configuration for single or multi-speaker speech
      oneOf:
      - type: object
        required:
        - voiceConfig
        properties:
          voiceConfig:
            $ref: '#/components/schemas/VoiceConfig'
            description: Single-speaker voice configuration
      - type: object
        required:
        - multiSpeakerVoiceConfig
        properties:
          multiSpeakerVoiceConfig:
            $ref: '#/components/schemas/MultiSpeakerVoiceConfig'
            description: Multi-speaker voice configuration (up to 2 speakers)
    SpeechGenerationRequest:
      type: object
      required:
      - contents
      - generationConfig
      properties:
        contents:
          type: array
          description: Text content to convert to speech
          items:
            $ref: '#/components/schemas/Content'
        generationConfig:
          $ref: '#/components/schemas/SpeechGenerationConfig'
          description: Configuration for speech generation
    SpeechGenerationResponse:
      type: object
      properties:
        candidates:
          type: array
          description: Generated audio candidates
          items:
            $ref: '#/components/schemas/AudioCandidate'
        usageMetadata:
          $ref: '#/components/schemas/UsageMetadata'
          description: Token usage information
    AudioContent:
      type: object
      properties:
        parts:
          type: array
          description: Audio data parts
          items:
            type: object
            properties:
              inlineData:
                type: object
                required:
                - mimeType
                - data
                properties:
                  mimeType:
                    type: string
                    description: MIME type of the audio (PCM format)
                    example: audio/pcm
                  data:
                    type: string
                    format: byte
                    description: Base64-encoded audio data (24kHz, 16-bit, mono PCM)
    SpeakerVoiceConfig:
      type: object
      required:
      - speaker
      - voiceConfig
      properties:
        speaker:
          type: string
          description: Name of the speaker (must match names used in prompt)
          example: Joe
        voiceConfig:
          $ref: '#/components/schemas/VoiceConfig'
          description: Voice configuration for this speaker
    VideoMetadata:
      type: object
      description: Metadata for customizing video processing
      properties:
        start_offset:
          type: string
          description: Start time offset for video clipping (format like "1250s")
          pattern: ^\d+s$
          example: 1250s
        end_offset:
          type: string
          description: End time offset for video clipping (format like "1570s")
          pattern: ^\d+s$
          example: 1570s
        fps:
          type: number
          format: float
          description: Custom frame rate sampling (frames per second). Default is 1 FPS. Use <1 for long videos, >1 for rapidly changing visuals
          minimum: 0.1
          example: 5
    FileData:
      type: object
      description: Reference to a file uploaded via the File API or a YouTube URL
      properties:
        mime_type:
          type: string
          description: MIME type of the file (optional for YouTube URLs)
          example: video/mp4
        file_uri:
          type: string
          description: URI of the file from File API or YouTube URL
          examples:
          - https://generativelanguage.googleapis.com/v1beta/files/abc123
          - https://www.youtube.com/watch?v=9hE5-98ZeCg
      required:
      - file_uri
    UsageMetadata:
      type: object
      properties:
        promptTokenCount:
          type: integer
          description: Number of tokens in the prompt
        candidatesTokenCount:
          type: integer
          description: Number of tokens in generated candidates
        totalTokenCount:
          type: integer
          description: Total token count
    MultiSpeakerVoiceConfig:
      type: object
      required:
      - speakerVoiceConfigs
      properties:
        speakerVoiceConfigs:
          type: array
          description: Voice configurations for each speaker (maximum 2)
          minItems: 2
          maxItems: 2
          items:
            $ref: '#/components/schemas/SpeakerVoiceConfig'
    InlineData:
      type: object
      required:
      - mime_type
      - data
      properties:
        mime_type:
          type: string
          description: MIME type of the data
          enum:
          - image/jpeg
          - image/png
          - image/webp
          - video/mp4
          - video/mpeg
          - video/mov
          - video/avi
          - video/x-flv
          - video/mpg
          - video/webm
          - video/wmv
          - video/3gpp
          - audio/wav
          - audio/mp3
          - audio/aiff
          - audio/aac
          - audio/ogg
          - audio/flac
          - application/pdf
        data:
          type: string
          format: byte
          description: Base64 encoded data
    AudioCandidate:
      type: object
      properties:
        content:
          $ref: '#/components/schemas/AudioContent'
          description: Generated audio content
        finishReason:
          type: string
          enum:
          - FINISH_REASON_UNSPECIFIED
          - STOP
          - MAX_TOKENS
          - SAFETY
          - RECITATION
          - OTHER
          description: Reason why generation stopped
        index:
          type: integer
          description: Index of the candidate
    PrebuiltVoiceConfig:
      type: object
      required:
      - voiceName
      properties:
        voiceName:
          type: string
          description: Name of the prebuilt voice with characteristic (Bright, Upbeat, Informative, etc.)
          enum:
          - Zephyr
          - Puck
          - Charon
          - Kore
          - Fenrir
          - Leda
          - Orus
          - Aoede
          - Callirrhoe
          - Autonoe
          - Enceladus
          - Iapetus
          - Umbriel
          - Algieba
          - Despina
          - Erinome
          - Algenib
          - Rasalgethi
          - Laomedeia
          - Achernar
          - Alnilam
          - Schedar
          - Gacrux
          - Pulcherrima
          - Achird
          - Zubenelgenubi
          - Vindemiatrix
          - Sadachbia
          - Sadaltager
          - Sulafat
          example: Kore
    Error:
      type: object
      properties:
        error:
          type: object
          properties:
            code:
              type: integer
              description: HTTP status code
            message:
              type: string
              description: Error message
            status:
              type: string
              description: Error status
    Part:
      type: object
      description: A part of the content, can be text, inline data, or file data
      oneOf:
      - type: object
        required:
        - text
        properties:
          text:
            type: string
            description: Text content
      - type: object
        required:
        - inline_data
        properties:
          inline_data:
            $ref: '#/components/schemas/InlineData'
      - type: object
        required:
        - file_data
        properties:
          file_data:
            $ref: '#/components/schemas/FileData'
          video_metadata:
            $ref: '#/components/schemas/VideoMetadata'
            description: Optional metadata for video processing (clipping, FPS)
    SpeechGenerationConfig:
      type: object
      required:
      - responseModalities
      - speechConfig
      properties:
        responseModalities:
          type: array
          description: Must include "AUDIO" for TTS
          items:
            type: string
            enum:
            - AUDIO
          example:
          - AUDIO
        speechConfig:
          $ref: '#/components/schemas/SpeechConfig'
          description: Configuration for speech output
    VoiceConfig:
      type: object
      required:
      - prebuiltVoiceConfig
      properties:
        prebuiltVoiceConfig:
          $ref: '#/components/schemas/PrebuiltVoiceConfig'
  securitySchemes:
    OAuth2:
      type: oauth2
      description: OAuth 2.0 authentication for Google Books API
      flows:
        authorizationCode:
          authorizationUrl: https://accounts.google.com/o/oauth2/v2/auth
          tokenUrl: https://oauth2.googleapis.com/token
          scopes:
            https://www.googleapis.com/auth/books: Manage your Google Books library