云创展汇

ENGLISH
ENGLISH
DEUTSCH
日本語
|
Sign In
Start

Solutions

About Us

Services

News Center  >

News Detail

Tencent Hunyuan has announced the open source release of its voice-driven digital human model, HunyuanVideo-Avatar.

Source:Internet
2025-06-27 07:05:55

Recently, Tencent Hunyuan has released and open-sourced its voice-driven digital human model, HunyuanVideo-Avatar.


It is reported that HunyuanVideo-Avatar was jointly developed based on Tencent Hunyuan's large video model (HunyuanVideo) and MuseV technology from Tencent Music's Tianqin Lab. It supports head-shoulder, half-body, and full-body shots, as well as multi-style, multi-species, and two-person scenarios, providing video creators with video generation capabilities featuring high consistency and high dynamism.


Users can upload a character image and audio, and the HunyuanVideo-Avatar model will automatically understand the image and audio—such as the character's surrounding environment and the emotions contained in the audio—enabling the character in the image to speak or sing naturally, and generating videos with natural expressions, lip-sync, and full-body movements.


Currently, it supports uploading audio of no more than 14 seconds for video generation, and other capabilities will be gradually launched and open-sourced in the future.

Other News

  • Tencent Hunyuan has announced the open source release of its voice-driven digital human model, HunyuanVideo-Avatar.

  • JD.com ushers in the "3D Information Flow Era" with the launch of autostereoscopic 3D ads, smart home configurator, and 3D live streaming

  • Capture Cloud 3D Cloud Investment Promotion: A New Engine for the Digital Transformation of Industrial Park Investment Attraction