Over the past few months I've been working on a new PyTorch & Tensorflow project, which makes use of the NSynth Dataset. The goal with this project is to be able to generate sounds based on a written description, similar to the famous DALL·E project, wich synthesises images from text. There's still much to be done, but recently I've completed the CNN classifier, which allows to categorise 23 different sounds with a pretty decent accuracy. I'm very excited by all the new things I need to learn to continue with this project, and will post here as soon as I have my first usable results!
Get notified on new releases