Giters

MandilKarki / capstone-music-generation

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Music Generation with RNN and VAE

This project aims to generate and continue music using various machine-learning models. Specifically, we utilize the Long Short-Term Memory (LSTM) model and the Variational Autoencoder (VAE) for music generation. Our LSTM architecture includes 1024 Recurrent Neural Network (RNN) units and is initialized with the Glorot uniform distribution. The input music tokens are first passed through a learnable embedding layer to create 256-dimensional vectors, which are then recursively fed into the LSTM layer.

This project is the master repository which contains frontend and and backend Firstly clone the project

The project folder structure

|-- music_frontend //frontend
|-- music_generation //backend
|-- .gitignore
|-- docker-compose.yml
|-- README.md

Please follow the procedure

For backend

git clone https://github.com/fagami1423/music_generation/tree/API-Integratioin

For frontend

git clone https://github.com/fagami1423/music_frontend

Run the following master run commands

docker compose up

For Frontend

http://localhost

For Backend

http://localhost:8000/docs
or 
http://localhost:8000/redoc

About