v2.0

lipku/LiveTalkingv2.0Mar 31, 2026by lipku

AI Summary

Performs a major architectural refactor by reorganizing code into specific directories, implementing a plugin architecture for models and transmission, and adding RTMP output support.

Key Highlights

Major code refactoring with plugin architecture for models and transmission
Code reorganization into `avatars` and `streamout` directories
Added RTMP output support for transmission methods
Added Alibaba Cloud QwenTTS

Breaking Changes

Code refactored to plugin architecture
Transmission methods reorganized into `streamout` directory
Code moved to `avatars` directory

New Features

Plugin architecture
RTMP output
Alibaba Cloud QwenTTS
Code reorganization

Full Release Notes

1，重构代码，数字人模型和音频特征代码移到avatars目录下
2，数字人模型、tts、传输方式改成plugin方式接入
3，传输方式整理成单独的类，放到streamout目录下，添加rtmp输出
4，数字人推理和回贴公共部分代码整合到BaseAvatar中，各模型只需要实现自己特有部分代码
5，音频特征的切分代码整合到BaseAsr中
6，tts添加阿里云qwentts