Skip to main content
Advanced Lip Sync drives a person’s lip movements in a video to match provided speech. It supports two input modes — synthesize speech from text, or supply your own audio file. This endpoint requires a session_id obtained from the Identify Face step. Always call Identify Face first.

​
Workflow overview

1. identify-face  →  session_id
2. advanced-lip-sync (session_id + speech input)  →  task_id
3. Poll GET /kling/v1/videos/advanced-lip-sync/{task_id}  →  video URL

​
Input modes

​
Text mode — built-in TTS

Provide text, voice_id, and voice_language. The platform synthesizes the audio using the specified voice and drives the lip movements. 
curl https://www.anyfast.ai/kling/v1/videos/advanced-lip-sync \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "input": {
      "session_id": "YOUR_SESSION_ID",
      "text": "Hello, welcome to my channel.",
      "voice_id": "girlfriend_1_cn",
      "voice_language": "en"
    }
  }'

​
Audio mode — custom audio file

Provide audio_url to drive lip movements directly from an existing audio recording. 
curl https://www.anyfast.ai/kling/v1/videos/advanced-lip-sync \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "input": {
      "session_id": "YOUR_SESSION_ID",
      "audio_url": "https://example.com/speech.mp3"
    }
  }'

​
Parameters

ParameterTypeRequiredDescription
input.session_idstringYesSession ID from the Identify Face step
input.face_image_urlstringNoReference face image URL for identity consistency
input.textstringText modeText for the character to speak
input.voice_idstringText modeVoice ID for TTS synthesis. See the Voice ID reference for available voices with audio previews.
input.voice_languagestringText modeLanguage code: zh or en
input.audio_urlstringAudio modePublic URL of an audio file

​
Polling

After the task is created, poll with GET /kling/v1/videos/advanced-lip-sync/{task_id} using the Task Query endpoint. Status transitions: queued → processing → succeeded / failed. On success, the video download URL is available in data.data.task_result.videos[0].url.

Prerequisites: Identify Face

You must call this first to obtain a session_id.

Voice ID Reference

Browse all available voice IDs with audio previews to choose the right voice for your lip sync.

API Reference

View the interactive API playground for Kling Advanced Lip Sync.