What genre is my song? Drop it in and find out.
Standby — waiting for a track
Your verdict appears right here.
Know exactly which category and genre
your music fits into.
You’ll get a detailed explanation of why the song fits in the chosen genre, along with its measured loudness and a breakdown of its characteristics.
If you’re planning on distributing new music, you’ll be forced to choose a genre, then in most cases a sub genre. Some of these lists have twenty-plus genre titles that just leave you there, scratching your head and ultimately guessing. On top of that, the list changes — genres are added and merged all the time. As music progresses, so do its labels. So don’t guess and have your distribution company put you in the wrong category.
This is a fun, practical tool that I built when I found myself releasing music through a popular distribution company who needed me to pinpoint every song’s genre. With electronic music, it isn’t that easy. So I built this app through the eyes of a musician and music lover, to help musicians and music lovers.
No sign ups, and I don’t keep your songs — the file is trimmed in your browser and only a short excerpt is ever sent to be listened to. The verdicts sit in a log only your browser can open, and you can erase all of it whenever you like.
— Eric L. (eNeOe)
Is it free, and do I need an account?
It is free and there are no accounts, no sign-up and no email address to hand over. There is a cap on how many tracks one visitor can analyse per day, which resets at midnight UTC, and the genre glossary stays open either way.
Do you keep my music?
No. Your original, full-resolution audio file never leaves your device. Instead, your browser securely samples three short windows from the track, compresses them into a tiny, low-fidelity mono excerpt, and sends only that 54-second excerpt to our AI partner (Google Gemini) for analysis. This excerpt is processed ephemerally and is completely discarded immediately after the genre is identified. We do not store your music, and your unreleased tracks are never used to train AI models. The only thing saved is the written verdict—the category, genre and the reasoning—which is stored in a private log that only your browser can open.
Which distributor’s genre list does it use?
None of them exactly, because none of them publish one. Distribution platforms build their genre menus from the stores they deliver to and change them over time, so the built-in reference is a researched superset. If your upload form is in front of you, paste its options into the tool and the verdict is mapped onto that exact list instead.
What if my track genuinely sits between two genres?
Then it will tell you so rather than pretend otherwise. You get a ranked list — primary, secondary and tertiary genre, each with the audible evidence for it — and an explicit caution, in words, when a track straddles two styles. There is no percentage attached, on purpose: a certainty figure generated by an AI model would look like a measurement without being one.
What file types can I use?
MP3, WAV, FLAC, M4A, AAC, OGG, Opus and WebM — anything your browser can decode. Length does not matter much, since the excerpt is sampled from across the track rather than taken from the beginning.
Does it just guess from the filename or the metadata?
No. It listens. Before naming anything it decides whether it is hearing people playing instruments or something sequenced, then describes the pace, drum pattern, bass behaviour and harmony, and only then matches that description against the taxonomy. You can read all of it in the tabs under the verdict. The loudness figures are different again: they are measured by your browser from the track itself, not described by the model.