openapi: 3.1.0
info:
  title: Cartesia Auth TTS API
  description: Cartesia is a real-time multimodal AI platform built around the Sonic text-to-speech and Ink streaming speech-to-text models. The REST API supports synchronous and streaming generation, voice management, voice cloning, and access-token issuance.
  version: '2024-11-13'
  contact:
    name: Cartesia
    url: https://docs.cartesia.ai
servers:
- url: https://api.cartesia.ai
  description: Production
security:
- ApiKeyAuth: []
- BearerAuth: []
tags:
- name: TTS
paths:
  /tts/bytes:
    post:
      tags:
      - TTS
      summary: Generate speech audio as bytes
      operationId: ttsBytes
      parameters:
      - $ref: '#/components/parameters/CartesiaVersion'
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/TtsRequest'
      responses:
        '200':
          description: Binary audio bytes
          content:
            audio/wav:
              schema:
                type: string
                format: binary
            audio/mpeg:
              schema:
                type: string
                format: binary
            application/octet-stream:
              schema:
                type: string
                format: binary
  /tts/sse:
    post:
      tags:
      - TTS
      summary: Generate speech audio as a Server-Sent Events stream
      operationId: ttsSse
      parameters:
      - $ref: '#/components/parameters/CartesiaVersion'
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/TtsRequest'
      responses:
        '200':
          description: SSE stream of audio chunks
          content:
            text/event-stream:
              schema:
                type: string
components:
  schemas:
    TtsRequest:
      type: object
      required:
      - model_id
      - transcript
      - voice
      - output_format
      properties:
        model_id:
          type: string
          example: sonic-2
        transcript:
          type: string
        voice:
          type: object
          properties:
            mode:
              type: string
              enum:
              - id
              - embedding
            id:
              type: string
            embedding:
              type: array
              items:
                type: number
        language:
          type: string
        output_format:
          type: object
          properties:
            container:
              type: string
              enum:
              - raw
              - wav
              - mp3
            sample_rate:
              type: integer
            encoding:
              type: string
        speed:
          type: number
        duration:
          type: number
  parameters:
    CartesiaVersion:
      in: header
      name: Cartesia-Version
      required: true
      schema:
        type: string
        example: '2024-11-13'
  securitySchemes:
    ApiKeyAuth:
      type: apiKey
      in: header
      name: X-API-Key
      description: Cartesia API key (sk_car_...)
    BearerAuth:
      type: http
      scheme: bearer
      description: Short-lived client access token