一种基于WebRTC的实时同步数字人音频与画面的方法

By leveraging the collaborative processing capabilities of WebRTC technology, the problem of asynchronous digital facial movements and audio data was resolved, achieving audio-visual synchronization and improving the stability and smoothness of the user experience.

CN117155903BActive Publication Date: 2026-07-17SHENZHEN RENDERBUS TECH
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SHENZHEN RENDERBUS TECH
Filing Date
2023-09-18
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

In existing WebRTC technology, digital facial movements and audio data are prone to desynchronization, which affects the user experience.

Method used

Through the collaboration of the web client, streaming server, application client, and backend server, audio data is directly transmitted and facial data is converted on the application client. Combined with a time-point correction mechanism, audio and video synchronization is ensured.

Benefits of technology

It achieves synchronization between digital facial movements and audio data, improving the stability and smoothness of the user experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117155903B_ABST
    Figure CN117155903B_ABST
Patent Text Reader

Abstract

本发明公开一种基于WebRTC的实时同步数字人音频与画面的方法。本发明通过Web端、推流服务器、应用端、后端服务器的高度协作，聚合了实时交互的流程，以及音频、脸部数据的处理，乃至整个实时交互流程，通过本策略，可以在跨终端的情况下，可以快速高效地建立起稳定的低时延实时交互数字人音画同步服务系统。
Need to check novelty before this filing date? Find Prior Art