Starstruk Logo MarkStarstruk

v2.0

clusterzx/paperless-aiv2.0Mar 31, 2026by lipku

AI Summary

Major refactoring of the digital avatar and audio feature code structure, migrating to a plugin-based architecture and adding RTMP output support.

Key Highlights

  • Refactored code structure moving models and audio features to `avatars` directory
  • Migrated digital avatar, TTS, and transport methods to plugin-based architecture
  • Added RTMP output support
  • Integrated common inference and audio feature code

New Features

  • Plugin-based architecture for avatar and audio components
  • RTMP output stream support
  • Consolidated common inference code into BaseAvatar
  • Consolidated audio feature splitting code into BaseAsr

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