v2.0

ekzhang/borev2.0Mar 31, 2026by lipku

AI Summary

A major refactor introducing a plugin-based architecture, reorganizing code into specific directories, and adding RTMP output support.

Key Highlights

  • Refactored code to move digital human models and audio features to the `avatars` directory.
  • Converted models and transmission methods to a plugin-based architecture.
  • Moved transmission classes to the `streamout` directory with RTMP output support.
  • Integrated common inference and response code into BaseAvatar.
  • Added Alibaba Cloud QwenTTS support.

Breaking Changes

  • Architecture refactoring to plugin-based models and transmission.
  • Code reorganization (avatars, streamout directories).

New Features

  • Plugin-based architecture
  • RTMP output support
  • Alibaba Cloud QwenTTS integration

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