feat: add model information and realtime factor to benchmark

fix: re-enable Blackwell GPU detection after PyTorch 2.9+ upgrade

Remove hardcoded HF token from scripts

Add H.264 WebSocket streaming and React.js web interface

Add complete MuseTalk real-time avatar system with WebSocket streaming

Add DemoTalk Benchmark for end-to-end latency testing

Add WebRTC streaming and demotalk automated test

Enhance metrics to capture all frame timings

Add metrics system for performance monitoring

Fix proxy compatibility and improve streaming

Add WebSocket streaming, idle animation, and sync improvements

Add speech-to-speech avatar server with real-time video generation

Fix compatibility with Python 3.12 and PyTorch 2.9+

Add avatar configuration and audio files

Add hello_world config and update project files

Update requirements.txt

Update scripts/inference_alpha.py

Add files using upload-large-folder tool

Add files using upload-large-folder tool

Add files using upload-large-folder tool

Add files using upload-large-folder tool

Add files using upload-large-folder tool

initial commit