Skip to main content
Face identification is the first step in the Kling lip sync workflow. It analyzes faces in a video and returns a session_id that must be used in the subsequent lip sync generation request.

​
Workflow Overview

identify-face  →  session_id  →  advanced-lip-sync  →  task_id  →  Poll for result
The session_id binds the video to the face analysis result. Without it, lip sync generation cannot proceed.

​
Video Requirements

  • The video must contain at least one clearly visible, front-facing face
  • Faces that are well-lit and without heavy obstruction produce better results
  • The video must be accessible via a public URL, or use a previously generated Kling video ID (video_id)

​
Quick Example

curl https://ai.alad.com/kling/v1/videos/identify-face \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "video_url": "https://example.com/person-speaking.mp4"
  }'

​
Request Parameters

ParameterTypeRequiredDescription
video_urlstringOne of the twoPublic URL of the source video
video_idstringOne of the twoID of a previously generated Kling video

​
Response

This endpoint is synchronous — results are returned immediately with no polling needed.
FieldDescription
data.session_idSession ID to pass to the lip sync endpoint
data.face_list[]All faces detected in the video
face_list[].face_idUnique identifier for each face
face_list[].face_rectFace bounding box {x, y, width, height} (pixels)
When multiple people are in the video, face_list will contain multiple entries.

Next Step: Lip Sync Generation

Use the session_id to generate a lip sync video.

API Reference

View the interactive API documentation for Kling Face Identification.