v2.0
CL-lau/SQL-GPTv2.0Mar 31, 2026by lipku
AI Summary
A major refactoring that reorganizes the codebase by moving digital human models and audio features into dedicated directories. It introduces a plugin-based architecture for models and TTS, integrates RTMP streaming support, and adds a new Aliyun QwenTTS provider.
Key Highlights
- Refactored code to move digital human models and audio features into the avatars directory
- Switched digital human models and TTS to a plugin-based architecture
- Added RTMP output support by organizing transmission methods into a separate streamout class
- Integrated Aliyun QwenTTS as a new TTS provider
New Features
- Aliyun QwenTTS support
- RTMP output support
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