zak-45 / WLEDAudioSyncRTMGC

Real Time Music Genre Classification using Essentia.js

Music autotagging with Discogs Effnet

Realtime music autotagging based on the Discogs taxonomy (400 subgenre tags). Sending topTags via OSC.

This is a 'hack' of the actual Essentia.js demo :

https://github.com/MTG/essentia.js/tree/819c00e8f034dc709e3d330ed8ac9ed37dfdb767/examples/demos/discogs-autotagging

- Replaced YT by audio device capture.
- Should work on SmartPhone, Tablet,  PC computer ... every modern device with Chromium Based Browser.
- Send top 5 predictions to OSC server via UDP. 
- Designed for easy integration to WLEDAudioSync Chataigne Module but should work with any OSC server.

Node.js version used : 18.6. Dev made with vue.js & vite.js. Using osc.js (2.4.4) for OSC communication.

Easy start :

https://github.com/zak-45/WLEDAudioSyncRTMGC/releases

Take your binary from release file depend of your OS, run it ! 
This is a portable nodejs version with all included (made with pkg). 

Once running, open your browser and go to https://localhost:8000/WLEDAudioSyncRTMGC/

You need to accept the self-signed certificate. 
If you want to customize it, download the source, generate a certificate 
and copy the .key and .cert file under security folder.

To start project locally (Dev version):

• download all to your nodejs work folder

• cd server

• npm install for the first time

• npm run dev

• cd views
• npm install for the first time
• npm run dev

To build for deployment:

• cd views
• npm run build

Nota:

Default server port : 8000
Default OSC ip address : 127.0.0.1
Default OSC port : 12000

Can be changed via Env variables.
Good to set ENV_NODE=Production for better, increased performance before run the app.