openapi: 3.0.3 info: title: FishAudio 文字对口型 version: 1.0.0 description: >- Import this file into a private Coze cloud plugin. It generates speech from text with a selected Fish Audio voice, then creates and queries an asynchronous lip-synced video job. Every user supplies their own Fish Audio API key through the Authorization tool parameter. servers: - url: https://fishaudio.org paths: /api/open/v1/media/video-dubbing/jobs: post: operationId: create_text_lip_sync summary: Generate speech and create a lip-synced video description: >- Generate TTS audio from text and a Fish Audio voice ID, then create one asynchronous lip-sync job. Save data.id and reuse the same Idempotency-Key only when retrying this exact request. parameters: - $ref: '#/components/parameters/Authorization' - $ref: '#/components/parameters/IdempotencyKey' requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/CreateTextLipSyncRequest' responses: '200': description: TTS completed and the lip-sync job was created successfully content: application/json: schema: $ref: '#/components/schemas/VideoDubbingResponse' '400': $ref: '#/components/responses/ErrorResponse' '401': $ref: '#/components/responses/ErrorResponse' '402': $ref: '#/components/responses/ErrorResponse' '404': $ref: '#/components/responses/ErrorResponse' '413': $ref: '#/components/responses/ErrorResponse' '415': $ref: '#/components/responses/ErrorResponse' '422': $ref: '#/components/responses/ErrorResponse' '500': $ref: '#/components/responses/ErrorResponse' /api/open/v1/media/video-dubbing/jobs/{jobId}: get: operationId: get_text_lip_sync_job summary: Query a text-to-lip-sync video job description: >- Query the original job ID. Polling does not generate speech again and does not create or charge for another lip-sync job. parameters: - $ref: '#/components/parameters/Authorization' - name: jobId in: path required: true description: The data.id returned by create_text_lip_sync. schema: type: string responses: '200': description: Job status returned successfully content: application/json: schema: $ref: '#/components/schemas/VideoDubbingResponse' '400': $ref: '#/components/responses/ErrorResponse' '401': $ref: '#/components/responses/ErrorResponse' '404': $ref: '#/components/responses/ErrorResponse' '500': $ref: '#/components/responses/ErrorResponse' components: parameters: Authorization: name: Authorization in: header required: true description: Customer Fish Audio API key in the format Bearer YOUR_API_KEY. schema: type: string example: Bearer YOUR_FISH_AUDIO_API_KEY IdempotencyKey: name: Idempotency-Key in: header required: true description: >- A unique client-generated value. Reuse it only to retry the same request; use a new value for every new video generation. schema: type: string maxLength: 200 example: coze-text-lip-sync-20260822-001 schemas: CreateTextLipSyncRequest: type: object required: - video_url - text - reference_id properties: video_url: type: string format: uri description: Customer-owned or authorized video URL downloadable by Fish Audio. text: type: string minLength: 1 maxLength: 5000 description: Text used to generate the replacement speech track. reference_id: type: string minLength: 1 description: Fish Audio system or private voice ID used for TTS. model_id: type: string default: fishaudio-s21pro-flash description: Fish Audio speech synthesis model. Use the default for the first test. enum: - fishaudio-s21pro - fishaudio-s21pro-flash - fishaudio-s2pro - fishaudio-s1 speed: type: number minimum: 0.5 maximum: 2 default: 1 description: Speech speed multiplier from 0.5 to 2. volume: type: number minimum: -20 maximum: 20 default: 0 description: Output volume adjustment from -20 to 20. pitch: type: number minimum: -12 maximum: 12 description: Optional pitch adjustment in semitones. stability: type: number minimum: 0.5 maximum: 1.5 description: Optional voice stability control from 0.5 to 1.5. similarity: type: number minimum: 0.5 maximum: 1.5 description: Optional voice similarity control from 0.5 to 1.5. language: type: string maxLength: 32 description: Optional language hint supported by the selected model. emotion: type: string maxLength: 64 description: Optional emotion supported by the selected model. instruction: type: string maxLength: 1600 description: Optional style, role, dialect, pacing, or emotion instruction. text_normalization: type: boolean default: true description: Normalize text before synthesis. VideoDubbingResponse: type: object required: - success - data properties: success: type: boolean message: type: string requestId: type: string data: type: object required: - id - status - stage - progress properties: id: type: string description: Save this value as jobId for subsequent queries. status: type: string enum: - pending - processing - completed - failed stage: type: string enum: - lip_sync - completed - failed progress: type: number minimum: 0 maximum: 100 result_url: type: string format: uri nullable: true description: Temporary generated video URL when the job is completed. audio_url: type: string format: uri nullable: true description: TTS audio URL generated during the first step. error_message: type: string nullable: true created_at: type: string updated_at: type: string completed_at: type: string nullable: true estimated_time: type: number nullable: true result_expires_at: type: string nullable: true result_expired: type: boolean credits_used: type: object properties: tts: type: number lip_sync: type: number total: type: number quota_remaining: type: number nullable: true idempotency_replayed: type: boolean Error: type: object required: - code - message properties: code: type: string message: type: string requestId: type: string additionalProperties: true responses: ErrorResponse: description: Fish Audio API error content: application/json: schema: $ref: '#/components/schemas/Error'