Generative Music

This introduction includes Music Seperationm, Music Generation, etc.


Music Seperation

Spleeter

Paper: Spleeter: A FAST AND STATE-OF-THE ART MUSIC SOURCE SEPARATION TOOL WITH PRE-TRAINED MODELS
Code: deezer/spleeter


Wave-U-Net

Paper: Wave-U-Net: A Multi-Scale Neural Network for End-to-End Audio Source Separation
Code: f90/Wave-U-Net


Hyper Wave-U-Net

Paper: Improving singing voice separation with the Wave-U-Net using Minimum Hyperspherical Energy
Code: jperezlapillo/hyper-wave-u-net
MHE regularisation:


Demucs

Paper: Music Source Separation in the Waveform Domain
Code: facebookresearch/demucs


Music Generation

OpenAI Jukebox

Blog: Jukebox
model modified from VQ-VAE-2 Paper: Jukebox: A Generative Model for Music
Colab: Interacting with Jukebox


DeepSinger

Blog: Microsoft’s AI generates voices that sing in Chinese and English
Paper: DeepSinger: Singing Voice Synthesis with Data Mined From the Web
Demo: DeepSinger: Singing Voice Synthesis with Data Mined From the Web

The alignment model based on the architecture of automatic speech recognition

The architecture of the singing model

The inference process of singing voice synthesis


MusicGen

Paper: Simple and Controllable Music Generation
Code: https://github.com/facebookresearch/audiocraft


Tiny Audio Diffusion

Code: https://github.com/crlandsc/tiny-audio-diffusion



This site was last updated November 15, 2024.