This site works great for EDM. For example, just lookup some random future house song and it will pump out tons of similar tracks. This is a great tool for people that want to keep music consistent in their content, but don't want it to get boring. I will be using this for as long as possible!
Show HN: I trained an AI model on 120M+ songs from iTunes
291–300 of 444 posts
Re: Show HN: I trained an AI model on 120M+ songs from iTunes
#292I’ve tried ‘Allan Holdsworth Sand’ and ‘Allan Holdsworth Non-brewed Condiment’ and results were completely unrelated, I would even say random. Then I’ve tried ‘Allan Holdsworth Sixteen Men of Tain’ and at least it found some jazz tracks. I guess a track needs to have a simple static rhythm to not throw off the algorithm.
Re: Show HN: I trained an AI model on 120M+ songs from iTunes
#293Umm... is it just me or there's no "Search" button? Linux, Firefocks 50.1.
I definitely have to improve the UI, but for now, as you type in a song in the search bar, it should load some auto-complete suggestions that you can choose from.
Re: Show HN: I trained an AI model on 120M+ songs from iTunes
#294This is amazing. I actually really like that it gives you songs across completely different genres and moods (while keeping the beat). One piece of feedback - it'd be nice if the search box didn't clear if you click out of it to e.g. look at another tab while it loads.
If you ever write a blog post about the process of making this, I'd love to read it.
Re: Show HN: I trained an AI model on 120M+ songs from iTunes
#295Feedback: "sounds like" for music is more than just rhythm and tone. The search results are all useless for me because two songs with similar sheet music often have wildly different lyrical styles and subjects.
What would you say was the best music similarity search or recommendation engine you've used?
Re: Show HN: I trained an AI model on 120M+ songs from iTunes
#296I have been wanting to do similar things. But, shied away because I wasn't in the mood to find out if this sort of thing was legal or not. Can someone here who knows this space better talk about the legal aspect of doing something like this?
You're just interacting with the API, you're not gonna have any legal issues. I have some experience with Spotify's API (not MusicKit), so I'm gonna try guessing how it works based on that.
There's an API endpoint called audio_features[0] that tells you things about the song (tempo, danceability, acousticness, major/minor key...), so while you can't get full versions of every song, you can approximate how they sound like based on Spotify's audio analysis of them.
So, build a database of audio_features while respecting API limits, find the most similar ones based on about a dozen variables, and you're good to go.
[0] https://developer.spotify.com/documentation/web-api/referenc...
Re: Show HN: I trained an AI model on 120M+ songs from iTunes
#297They'll either be sued into the ground or bought out for a multi-million figure by the Summer!
Good luck!
Re: Show HN: I trained an AI model on 120M+ songs from iTunes
#298Earlier quoted context omitted.
I guess he is working with the 30s low rez previews, I know you can download them with the Spotify API. Apple Music should be similar.
Wouldn't this still be about 36 terabytes of data?
You would only need to store the embeddings and not the file audio itself.
Re: Show HN: I trained an AI model on 120M+ songs from iTunes
#299It doesn’t appear to work in Firefox. Looks like CORS is blocking a call to segment.
Re: Show HN: I trained an AI model on 120M+ songs from iTunes
#300This is very interesting, but unfortunately I haven't had the greatest luck in finding new songs I would enjoy listening to. It absolutely finds similar sounding tracks, but it doesn't distinguish which part of the song made it enjoyable. There's no tempo consistency or genre consistency or even main instrument/vocal timbre consistency between recommendations. I think locking one or more of those dimensions would all…
Categorising music is surprisingly different.
See this paper from https://everynoise.com/ : https://everynoise.com/EverynoiseIntro.pdf
IIRC they try to classify music on 17 different points/features. What you see on the web is an attenpt to visualise (and provide a guide to music based on) some of them