ElevenLabs Voice Cloning API

openapi: 3.1.0
info:
  title: ElevenLabs Voice Cloning API
  description: >-
    The ElevenLabs Voice Cloning API allows developers to create custom AI
    voices from audio recordings. Instant Voice Cloning requires as little as
    60 seconds of clean audio to generate a usable voice clone, while
    Professional Voice Cloning produces higher fidelity results from a minimum
    of 30 minutes of recordings. Cloned voices can then be used with the Text
    to Speech API for generating speech that closely matches the original
    speaker.
  version: '1.0'
  contact:
    name: ElevenLabs Support
    url: https://help.elevenlabs.io
  termsOfService: https://elevenlabs.io/terms-of-service
externalDocs:
  description: ElevenLabs Voice Cloning API Documentation
  url: https://elevenlabs.io/docs/api-reference/voices/ivc/create
servers:
  - url: https://api.elevenlabs.io
    description: Production Server
tags:
  - name: Instant Voice Cloning
    description: >-
      Endpoints for creating voice clones from short audio samples with
      instant processing.
  - name: Professional Voice Cloning
    description: >-
      Endpoints for creating high-fidelity voice clones from longer audio
      recordings with professional-grade processing.
security:
  - apiKeyAuth: []
paths:
  /v1/voices/add:
    post:
      operationId: createInstantVoiceClone
      summary: Create instant voice clone
      description: >-
        Creates a new voice clone from uploaded audio samples using instant
        voice cloning. Requires a minimum of 60 seconds of clean audio. The
        cloned voice is immediately available for use with speech generation
        endpoints.
      tags:
        - Instant Voice Cloning
      requestBody:
        required: true
        content:
          multipart/form-data:
            schema:
              $ref: '#/components/schemas/InstantVoiceCloneRequest'
      responses:
        '200':
          description: Voice clone created successfully
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/VoiceCloneResponse'
        '400':
          description: Bad request - invalid audio or parameters
        '401':
          description: Unauthorized - invalid or missing API key
        '422':
          description: Unprocessable entity - audio too short or low quality
  /v1/voices/{voice_id}/professional:
    post:
      operationId: createProfessionalVoiceClone
      summary: Create professional voice clone
      description: >-
        Initiates professional voice cloning from uploaded audio samples.
        Requires a minimum of 30 minutes of high-quality recordings. The
        cloning process takes longer than instant cloning but produces
        higher fidelity results.
      tags:
        - Professional Voice Cloning
      parameters:
        - $ref: '#/components/parameters/voiceId'
      requestBody:
        required: true
        content:
          multipart/form-data:
            schema:
              $ref: '#/components/schemas/ProfessionalVoiceCloneRequest'
      responses:
        '200':
          description: Professional voice clone initiated successfully
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/VoiceCloneResponse'
        '400':
          description: Bad request - invalid audio or parameters
        '401':
          description: Unauthorized - invalid or missing API key
        '422':
          description: Unprocessable entity - insufficient audio quality or duration
components:
  securitySchemes:
    apiKeyAuth:
      type: apiKey
      in: header
      name: xi-api-key
      description: >-
        ElevenLabs API key passed in the xi-api-key header for authentication.
  parameters:
    voiceId:
      name: voice_id
      in: path
      required: true
      description: >-
        The identifier of the voice to apply professional cloning to.
      schema:
        type: string
  schemas:
    InstantVoiceCloneRequest:
      type: object
      required:
        - name
        - files
      properties:
        name:
          type: string
          description: >-
            The name for the cloned voice.
        description:
          type: string
          description: >-
            Description of the cloned voice and its intended use.
        labels:
          type: string
          description: >-
            JSON string of key-value label pairs describing the voice
            characteristics such as accent, age, and gender.
        files:
          type: array
          description: >-
            Audio sample files for cloning. A minimum of 60 seconds of
            clean audio is recommended for best results.
          items:
            type: string
            format: binary
    ProfessionalVoiceCloneRequest:
      type: object
      required:
        - files
      properties:
        files:
          type: array
          description: >-
            High-quality audio recordings for professional voice cloning.
            A minimum of 30 minutes of recordings is required.
          items:
            type: string
            format: binary
        consent:
          type: string
          format: binary
          description: >-
            A signed consent form or audio consent from the voice owner
            authorizing the creation of the voice clone.
    VoiceCloneResponse:
      type: object
      properties:
        voice_id:
          type: string
          description: >-
            The unique identifier of the newly created voice clone.