song-generation 开源项目推荐（32 个）

AaronZ345 /VersBand

PyTorch Implementation of VersBand(EMNLP 2025): Versatile Framework for Song Generation with Prompt-based Control

scragnog /HOT-Step-CPP

转动旋钮。召唤神曲！现在有更多 C++！由 GGML 驱动的本地 AI 音乐生成

declare-lab /jamify

JAM：一款基于微型流程的歌曲生成器，具有细粒度的可控性和美学一致性

op0ai /wav0

AI-Native Studio for everything audio. Sound Design/Audio Editing/AI Producer/Fast Exports. Built for web, runs anywhere.

AlekseyKorshuk /huggingartists

Lyrics generation with GPT2-based Transformer

igrbible /Ruach_Studio

YuE2 Music Smithery in Ruach Studio: YuE2 Songs Creator, AI Writer, Refiner (30+ nodes, Piano Roll, DAW), Cover Artist, MiniMax H3 Videographer, LoRA Trainer, Librarian. SUNO AI alternative.

6Morpheus6 /wan2gp

Unified wan2gp installer for NVIDIA, AMD and Apple Mac, with Python 311, Pytorch 2.10 and NVFP4 Support for RTX50. Super Optimized Gradio UI for AI video creation (6GB+ VRAM). Wan 2.1/2.2, Qwen, Hunyuan Video, LTX Video, Flux and OVI and more. (AMD Windows supported by 7900(XT), 7800(XT), 7600(XT), Phoenix, 9070(XT) and Strix Halo)

Anil-matcha /suno-comfyui

ComfyUI custom nodes for Suno — generate, remix, extend, and shape AI music inside ComfyUI via the muapi.ai API.

Anil-matcha /awesome-minimax-music-3-prompts

Curated MiniMax Music 3.0 prompt library, lyrics-formatting guide, and MuAPI Python client for text-to-music generation

MASSIVEMAGNETICS /SUNOKILLER

SUNOKILLER is an advanced audio synthesis system that generates high-quality, human-sounding music and singing voices using state-of-the-art neural networks. Designed to run efficiently on low-end hardware while delivering unmatched quality.

JustinValentine /DECtalk_RNN

DECtalk song generator using a simple RNN architecture

MASSIVEMAGNETICS /Song-Bloom-Bando-fied-Edition

This repository features the revolutionary Next-Gen X3 upgrade with Cognitive Architecture - moving beyond passive RAG to holographic, hyperdimensional computing:

Jdka1 /Drake-AI

Researching AI imitations in songwriting for artist cloning safety 🎤

Anil-matcha /minimax-music-3-comfyui

ComfyUI custom nodes for MiniMax Music 3.0 text-to-music generation via MuAPI

MindrLabs /music-generation-yue2

Turn lyrics and a style prompt into a full 48 kHz song and its ABC score, locally with YuE2-3B and model-compose.

asigalov61 /Karaoke-MASTER

Very efficient and cute GPT2-based Karaoke Melody Music AI creator and generator :) Konnichiwa !!! :)

theelderemo /genius-lyrics-cleaned-dataset

3.17M cleaned & deduplicated English song lyrics from Genius: title, artist, genre tag, year, and full lyrics. Mirror of HuggingFace dataset. MIT licensed.

siliconsense /yue2-studio-pc

One batch file that runs YuE2 on your own NVIDIA card: songs from lyrics, and covers sung to a melody transcribed from your own recording. Windows, 6 GB and up, roughly real time.

SamurAIGPT /minimax-music-3-api

Python wrapper for MuAPI's MiniMax Music 3.0 text-to-music API

useapi /flowmusic-api

Google Flow Music API examples by useapi.net: a Lyria API for full songs with Lyria 3.5 and Lyria 3 Pro (vocals, custom lyrics, instrumentals, cover, extend, stems) over REST on your Flow Music plan, or the one included with Google AI Plus, Pro or Ultra, plus a hosted MCP server (18 tools).