Skip to content

lyramakesmusic/sample-browser

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

40 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

sample-browser

sound-similarity sample browser via CLAP embeddings cosine distance

requires ~4.2 GB VRAM to run (you will need a GPU)

Install:

download this repo, unzip it into a folder, and run install.bat. you will need CUDA and python installed.

to run after first install, run run.bat. it will open localhost for you

code is multiplatform, but installer and run file are windows-only.

other OSs: create a venv python -m venv venv, enter it source venv/Scripts/activate, run pip install flask laion_clap librosa numpy torch, make sure the CUDA version of torch is installed pip install -U torch==2.4.0+cu121 torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121, then run python sound-similarity-browser.py and go to http://localhost:5000/

Usage:

Paste a local filepath into the Cache Management input and press Process Folder. When complete, upload a sound or type a sound description and press search

screenshot of sample browser showing a list of matching samples

screenshot of sample browser showing a progress bar caching audio latents

About

sound-similarity sample browser via clap

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published