All Products
Search
Document Center

Alibaba Cloud Model Studio:Lip-sync replacement for videos - VideoRetalk

Last Updated:Oct 29, 2025

VideoRetalk is a video generation model that uses a character video and an audio file to generate a new video in which the character's lip movements are synchronized with the audio.

Important

This document applies to China (Beijing). To use the model, you must use an API key from the China (Beijing) region.

Model overview

Example

Input example

Output example

Character video:

Voice audio:

Billing and rate limits

Model

Unit price

RPS limit for task submission

Number of concurrent tasks

videoretalk

Pay-as-you-go, by the duration of the generated video: $0.011469/second

1

1

(At any given time, only one task is running. Other tasks in the queue are waiting.)

To request an increase in the records per second (RPS) limit for the model, send an email to modelstudio@service.aliyun.com. Your email must include your Alibaba Cloud account ID, the model, and the required RPS.

Calling the model

  • VideoRetalk is available on a pay-as-you-go basis. You can use this model only by making API calls. It cannot be tested in the Alibaba Cloud Model Studio console.

  • You can call the VideoRetalk model and provide a clear, front-facing video of a character and a clear audio file to generate a video with lip-sync replacement. For more information, see VideoRetalk video generation.