openapi: 3.0.1
info:
  title: WellSaid Labs Clips Text-to-Speech API
  description: REST API for WellSaid Labs AI text-to-speech. Render text to speech using studio-quality voice avatars via synchronous clip creation, low-latency audio streaming, and word-timing renders with subtitles. Authentication is performed with an X-Api-Key request header.
  termsOfService: https://wellsaidlabs.com/terms/
  contact:
    name: WellSaid Labs Support
    url: https://docs.wellsaidlabs.com
  version: '1.0'
servers:
- url: https://api.wellsaidlabs.com/v1
security:
- ApiKeyAuth: []
tags:
- name: Text-to-Speech
  description: Render text to speech as clips or audio streams.
paths:
  /tts/stream:
    post:
      operationId: streamTextToSpeech
      tags:
      - Text-to-Speech
      summary: Render text to speech as an audio stream.
      description: Renders the supplied text to speech with the chosen voice avatar and streams the result back as an audio/mpeg (MP3) stream as it is produced.
      parameters:
      - name: X-Enable-SSML
        in: header
        required: false
        description: Enables limited SSML translation for the input text.
        schema:
          type: boolean
      - name: Accept
        in: header
        required: false
        description: Desired audio media type for the streamed response.
        schema:
          type: string
          default: audio/mpeg
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/RenderRequest'
      responses:
        '200':
          description: Audio stream from a successful text-to-speech render.
          content:
            audio/mpeg:
              schema:
                type: string
                format: binary
        '400':
          $ref: '#/components/responses/BadRequest'
        '401':
          $ref: '#/components/responses/Unauthorized'
        '429':
          $ref: '#/components/responses/TooManyRequests'
        '508':
          description: Render loop detected. Likely a result of unusual phrasing, characters, and/or punctuation in the input text.
  /tts/clips:
    post:
      operationId: createClip
      tags:
      - Text-to-Speech
      summary: Create a new TTS clip asynchronously.
      description: Submits a text-to-speech render request and creates a clip that is produced asynchronously. Poll the clip endpoint to retrieve status and the rendered audio.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/RenderRequest'
      responses:
        '200':
          description: The created clip.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Clip'
        '400':
          $ref: '#/components/responses/BadRequest'
        '401':
          $ref: '#/components/responses/Unauthorized'
        '429':
          $ref: '#/components/responses/TooManyRequests'
  /tts/word-timing:
    post:
      operationId: renderWordTiming
      tags:
      - Text-to-Speech
      summary: Render text to speech with timing information and subtitles.
      description: Renders the supplied text to speech and returns a zip file containing the generated audio along with word-level timing as JSON, SRT, and VTT subtitle files.
      parameters:
      - name: X-Enable-SSML
        in: header
        required: false
        description: Enables limited SSML translation for the input text.
        schema:
          type: boolean
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/RenderRequest'
      responses:
        '200':
          description: Zip archive containing the rendered audio plus JSON, SRT, and VTT timing files.
          content:
            application/zip:
              schema:
                type: string
                format: binary
        '400':
          $ref: '#/components/responses/BadRequest'
        '401':
          $ref: '#/components/responses/Unauthorized'
        '429':
          $ref: '#/components/responses/TooManyRequests'
        '508':
          description: Render loop detected.
components:
  schemas:
    Clip:
      type: object
      properties:
        id:
          type: string
        status:
          type: string
          description: Processing status of the asynchronous render.
        speaker_id:
          type: integer
        text:
          type: string
        url:
          type: string
          description: URL of the rendered audio when available.
    RenderRequest:
      type: object
      required:
      - speaker_id
      - text
      properties:
        speaker_id:
          type: integer
          description: Identifier of the voice avatar (speaker) to render with.
          example: 50
        text:
          type: string
          minLength: 1
          maxLength: 1000
          description: The input text to render to speech.
          example: I love building and shipping new features for our users!
        model:
          type: string
          enum:
          - legacy
          - caruso
          description: The TTS model to render with.
        library_ids:
          type: array
          description: Optional replacement library identifiers to apply during the render.
          items:
            type: string
        audio_configs:
          type: object
          description: Optional audio configuration settings for the render.
          additionalProperties: true
  responses:
    BadRequest:
      description: Request failed input validation.
    Unauthorized:
      description: API key missing or invalid.
    TooManyRequests:
      description: Rate limit exceeded.
  securitySchemes:
    ApiKeyAuth:
      type: apiKey
      in: header
      name: X-Api-Key