FindGenre uses AI to identify music across a wide range of genres from an audio upload or a YouTube or SoundCloud link. It provides a genre prediction with a confidence score and is especially strong at distinguishing closely related electronic subgenres like Melodic Techno, Progressive House, Tech House, and Trance as well as different genre's of rock. We are constantly improving the model to provide more accurate results.

I built FindGenre because I was constantly running into tracks that were difficult to classify—especially within electronic music and rock, where the lines between genres and subgenres can be subjective. Existing tools were often too broad or inaccurate. I wanted to create a simple AI-powered tool that helps producers, DJs, labels, and music fans identify genres more consistently and better understand where a track fits. Training the model required an extensive library of 15000 tracks and some genre's had to be classified by actually listening to the music to limit genre bleed as I call it. On a technical level, I wanted to leverage my career in DevOps to separate FIndGenre into distinct microservices with the actual genre classifier a serverless gpu instance group which makes classification faster and is cost effective.

how well does it handle tracks that blend multiple genres, like something halfway between progressive house and melodic techno, does the confidence score reflect that mix or just pick one?

Curious how it handles remixes or mashups that blend two genres, does it just pick the dominant one or give you a split breakdown?

how well does it handle tracks that blend multiple genres, like something halfway between progressive house and melodic techno, does the confidence score reflect that mix or just pick one?

Curious how it handles remixes or mashups that blend two genres, does it just pick the dominant one or give you a split breakdown?

FindGenre

Find your track's genre with AI audio analysis

Find your track's genre with AI audio analysis