Xingchen Song(宋星辰)

S3Tokenizer. Reverse Engineering of Supervised Semantic Speech Tokenizer (S3Tokenizer) proposed in CosyVoice

FlashCosyVoice. FlashCosyVoice: A lightweight vLLM implementation built from scratch for CosyVoice.

TouchNet. A native-PyTorch library for large scale M-LLM (text/audio) training with tp/cp/dp.

Speech-Transformer-tf2.0. transformer for ASR-systerm (via tensorflow2.0)

CosyVoice-ttsfrd. Python

Speech-Transformer-plus-2DAttention. A PyTorch implementation of Speech Transformer, an End-to-End ASR with Transformer network on Mandarin Chinese.

ASR-Wavnet. some ASR-system implementations （via tensorflow 1.x）

dotfiles. my settings for dotfiles

ModulesPack2. ModulesPack version2

whisper.cpp. Port of OpenAI's Whisper model in C/C++

playground. happy developing

torchtitan. A PyTorch native library for large model training

OmniQuant. OmniQuant is a simple and powerful quantization technique for LLMs.

The-C-Programming-Language-Homework. My personal homework

bibfiles. my bibtex for paper reference

InferLLM. a lightweight LLM model inference framework

xingchensong. Config files for my GitHub profile.