Inworld AI TTS API

openapi: 3.1.0
info:
  title: Inworld AI Models TTS API
  description: 'REST API for Inworld AI''s voice and conversational AI platform, covering

    Text-to-Speech (synchronous and streaming), Speech-to-Text, voice

    management, models, and the LLM Router that fronts 220+ third-party

    models via the OpenAI-compatible chat completions endpoint.

    '
  version: '1.0'
  contact:
    name: Inworld AI
    url: https://docs.inworld.ai
servers:
- url: https://api.inworld.ai
  description: Production API
security:
- basicAuth: []
tags:
- name: TTS
  description: Text-to-Speech synthesis
paths:
  /v1/tts/synthesize:
    post:
      tags:
      - TTS
      summary: Synthesize speech
      description: Send text to the TTS REST API and receive a complete audio file in one response.
      operationId: synthesizeSpeech
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/SynthesizeRequest'
      responses:
        '200':
          description: Synthesized audio
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/SynthesizeResponse'
        '400':
          $ref: '#/components/responses/BadRequest'
        '401':
          $ref: '#/components/responses/Unauthorized'
  /v1/tts/synthesize:stream:
    post:
      tags:
      - TTS
      summary: Synthesize speech (streaming)
      description: Stream synthesized audio chunks over HTTP as they are processed.
      operationId: synthesizeSpeechStream
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/SynthesizeRequest'
      responses:
        '200':
          description: Streamed audio chunks (newline-delimited JSON)
          content:
            application/x-ndjson:
              schema:
                $ref: '#/components/schemas/SynthesizeResponse'
  /v1/tts/voices:
    get:
      tags:
      - TTS
      summary: List available TTS voices
      operationId: listTtsVoices
      responses:
        '200':
          description: A list of TTS voices
          content:
            application/json:
              schema:
                type: object
                properties:
                  voices:
                    type: array
                    items:
                      $ref: '#/components/schemas/Voice'
components:
  responses:
    Unauthorized:
      description: Missing or invalid API key
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/Error'
    BadRequest:
      description: Invalid request
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/Error'
  schemas:
    AudioConfig:
      type: object
      properties:
        audioEncoding:
          type: string
          enum:
          - LINEAR16
          - MP3
          - OGG_OPUS
          - ALAW
          - MULAW
          - FLAC
          - PCM
          - WAV
        sampleRateHertz:
          type: integer
          minimum: 8000
          maximum: 48000
        speakingRate:
          type: number
          minimum: 0.5
          maximum: 1.5
        bitRate:
          type: integer
          default: 128000
    SynthesizeResponse:
      type: object
      properties:
        audioContent:
          type: string
          format: byte
          description: Base64-encoded audio bytes in the requested format.
        usage:
          type: object
          properties:
            characterCount:
              type: integer
            modelId:
              type: string
        timestampInfo:
          type: array
          items:
            type: object
            properties:
              text:
                type: string
              startSeconds:
                type: number
              endSeconds:
                type: number
              phonemes:
                type: array
                items:
                  type: string
    Error:
      type: object
      properties:
        code:
          type: integer
        message:
          type: string
        status:
          type: string
    Voice:
      type: object
      properties:
        id:
          type: string
        name:
          type: string
        language:
          type: string
        gender:
          type: string
        description:
          type: string
        previewUrl:
          type: string
          format: uri
    SynthesizeRequest:
      type: object
      required:
      - text
      - voiceId
      - modelId
      properties:
        text:
          type: string
          maxLength: 2000
          description: Input string to convert to speech.
        voiceId:
          type: string
          description: Identifier for the desired voice.
        modelId:
          type: string
          example: inworld-tts-2
          description: TTS model selection (e.g., inworld-tts-2, inworld-tts-1.5-mini).
        audioConfig:
          $ref: '#/components/schemas/AudioConfig'
        language:
          type: string
          example: en-US
          description: BCP-47 language tag.
        deliveryMode:
          type: string
          enum:
          - STABLE
          - BALANCED
          - CREATIVE
          description: Output variation control (TTS-2 only).
        timestampType:
          type: string
          enum:
          - WORD
          - CHARACTER
        applyTextNormalization:
          type: string
          enum:
          - true
          - false
  securitySchemes:
    basicAuth:
      type: http
      scheme: basic
      description: 'Authentication uses HTTP Basic with the API key as the username

        (password empty), encoded as: `Authorization: Basic base64(apiKey:)`.

        Get an API key at https://platform.inworld.ai.

        '