0.5.1

What's Changed

• LiteAvatar support multi-session.
• Add support for the Qwen-Omni multimodal model, using the Qwen-Omni-Realtime API service of BaiLian.
• Modified to a separated frontend and backend version.
• Fix many bugs.

Contributors

@sudowind @lovepope @bingochaos @djiayong5 @HaveAnApplePie @raidios

Full Changelog: 0.4.0...0.5.1

0.4.1

What's Changed

• add MuseTalk support by @lovepope and @HaveAnApplePie

0.3.0

What's Changed

• Added support for LAM in digital humans, enabling concurrent configuration when LAM is selected. TTS now supports edge_tts and BaiLian CosyVoice.
• Updated dependency management approach based on UV and handler modules, supporting direct execution or using Docker.
• Added subtitle display on the front end.

0.2.2

• support change avatar
• use gpu as avatar default backend

0.2.1

• Added support for history
• Support for text input
• Camera requirement removed at startup
• Optimized modular loading method

v0.2.0

[feat] support minicpm-o-2.6-in4 in docker image, clone and build AutoGPTQ during startup. (#10)
[feat] added video modality input to MiniCPM
[feat] added tri-phase handlers(ASR+LLM+TTS)
[feat] restructured all handlers, handler related submodule and codes are moved to
the same folder alongside the main handler file.