donkeyhang Goto Github PK

audio-inpainting-diffusion

声音修复基于diffusion

audiofile

A simple C++ library for reading and writing audio files.

audiogpt

AudioGPT: Understanding and Generating Speech, Music, Sound, and Talking Head

chatgpt-next-web

One-Click to deploy well-designed ChatGPT web UI on Vercel. 一键拥有你自己的 ChatGPT 网页服务。

chatlm-mini-chinese

中文对话0.2B小模型（ChatLM-Chinese-0.2B），开源所有数据集来源、数据清洗、tokenizer训练、模型预训练、SFT指令微调、RLHF优化等流程的全部代码。支持下游任务sft微调，给出三元组信息抽取微调示例。

chatpaper

Use ChatGPT to summarize the arXiv papers. 全流程加速科研，利用chatgpt进行论文总结+润色+审稿+审稿回复

chinese-llama-2-7b

开源社区第一个能下载、能运行的中文 LLaMA2 模型！

chinese-llama-alpaca

中文LLaMA&Alpaca大语言模型+本地部署 (Chinese LLaMA & Alpaca LLMs)

colorsplitter

A cli tool for split vocal timbre.

deepfilternet

Noise supression using deep filtering

deeplearning-500-questions

深度学习500问

fish-speech

Brand new TTS solution

fucking-algorithm

刷算法全靠套路

generative-ai-for-beginners

Get Started Building with Generative AI

gpt-sovits

1 min voice data can also be used to train a good TTS model! (few shot voice cloning)

gpt-sovits-server

android gpt_sovits

gpt_sovits_cmdline

gptsovits commandline

grok-1

Grok open release

juce

JUCE is an open-source cross-platform C++ application framework for desktop and mobile applications, including VST, VST3, AU, AUv3, RTAS and AAX audio plug-ins.

juce_template

juce_template_cmake

kuiperinfer

从零实现一个高性能的深度学习推理库

mediapipe

厉害 Cross-platform, customizable ML solutions for live and streaming media.

music_ai_101

一些music资料

nemo

NeMo: a toolkit for conversational AI

no-name-svc

4G GPU & 10 Minutes for train

pamplejuce

pesto

pitch easimate lightning

retrieval-based-voice-conversion-new

Voice data <= 10 mins can also be used to train a good VC model!

sensevoice

语音克隆清洗数据、Multilingual Voice Understanding Model