Comment by fructose | Hacker News Reader

Comment by fructose | Hacker News Reader

I am working on a new music platform. It is a travesty that SoundCloud and other artist-friendly platforms like Bandcamp are struggling and can't survive. Having used Spotify, Tidal, Apple Music (iTunes), YouTube Music, Amazon Music, Deezer, Qobuz, etc over the past 20+ years, it seems like without a new approach we'll continue to lose choices in how we search for, discover, listen to, and interface with music - I can't tell you how many times I've ended up in dead-end-inducing discovery and recommendation experiences in the aformentioned services. I'm not optimistic that the big players in digital streaming can get out of their own way and do anything to make stark improvements to the way listeners experience music, let alone provide anything new to artists, unsigned artists, labels, playlist creators, DJ's, or people who own and operate music venues that aren't owned or controlled by giant corporations. And this is at a time when there is more music being created, recorded, and released than ever before (120,000 new tracks get published a day and 43 million new tracks were published in 2023 alone)(1).

The music project that I have been working on aims to bring together a musical audio pre-trained language model, natural language inputs, and a novel interface that deconstructs what you hear into visualizations and user controls. The idea is to allow you to dramatically improve personalization of music search, discovery, and curation by allowing interactivity with the individual parts of music, also known as stems in music theory and music mastering.

The overarching challenge that music streaming services and more broadly media streaming services have, is building a sustainable business on somebody else's art...you have to (rightly) pay licensing costs. This is a major (and ever changing) business risk and one of the larger challenges when I think about launching a new music streaming platform. MTV discovered this in the mid 80's when margins for music and music video licensing became flat, Netflix / Apple / Amazon / etc know this which is why they invested in developing their own content in house.

If anybody is passionate about working on something new in streaming music let me know - I could really use help developing the music audio PLM prototype.

(1) https://www.musicbusinessworldwide.com/there-are-now-120000-...

Noble goal and one I considered for a while after Streamus got shut down (https://thenextweb.com/news/how-youtube-killed-an-extension-...). I got to the point where I spoke with UMG on the phone (prick started the call talking about his Porsche) and Monstercat (they're cool in my book) but no further before YouTube's lawyers crushed me.

Yes, sustainable business model is hard. Just look at Spotify's stock - it's basically break even since IPO. The music industry adjusts their fees to keep Spotify alive, but never significantly profitable, and Spotify has little ability to push back on this.

...but there's also soooooooooo much entrenchment going on in this space. There's special contracts written that whitelist which platforms tracks can be played on and tricky "favored nation" clauses (https://www.nytimes.com/2015/05/25/business/media/sony-terms....) So, if you create a new platform, and it's a real threat to the others, then you just get denied a seat at the table unless you make concessions to the point of impotency... or you only have indie music that struggles to attract masses.

Ultimately I decided that, while I cared a LOT about this sector, that it's just too much of an uphill battle.

Your idea sounds cool though. I wish you luck. Is it your intent to build a music catalog that is entirely unique to your platform, to bring in indie artists, or to try and bring in the big names, too? If you're not trying to bring in the big names... how do you prevent the situation where someone says, "I love this song, but next I want to hear ____.", it's not on your platform, and so they open Spotify/YT, and now suddenly you've lost engagement with your platform. I don't feel that consumers maintain playlists across multiple music services and bounce between them based on their current listening preferences. Do you?

I am mostly focused on getting the PLM working right now but have sketched out a pretty sizable product that has a sort of re-envisioning of what a musical community is such as better identity building tools for artists, listeners, labels, publishers (non-musicians such as playlist builders and DJ's) and a bunch of other fundamental things that I think are flat out missing from music streaming apps.

I've had two thoughts re: how to start and how to build an interesting music catalog without getting crushed by the lawyers and labels who are in cahoots with each other. First, the concept of breaking music into its stems and letting people directly interface with music "particles" that their ear-brains love most has made me think that the most important thing to get right is recommending more music that has the sonic characteristics that someone enjoys most, not necessarily broad things that come from music metadata like genre, lyrics, or what your friends liked or listened to. Rather, things like melody, groove or rhythmic patterns, BPM, timbre, harmonics, textures, or music structural preferences are how you get recommendations on what to listen to next. When thinking about it that way, it seems like how the music sounds can be used to recommend, regardless of who big or small or new or established or signed or unsigned an artist might be. In my mind, there's an interesting democratic nature to that too where new artists get a fair shake at discovery as much as major artists (whereas today there's a lot of editorial control over who gets bubbled up into recommendations algos). Second, I've thought about developing a solid prototype and then going for sizeable investment so that I could afford to strike deals, join the cabal, and establish the relationships and access the catalogs of labels big and small.

Would love to know more about this project.

I been working on UX with the team at https://www.radiooooo.com and I believe there is a lot of space for new UI and discovery interactions that treat finding new music as a game.

There is an unspeakable ‘uncanny valley‘ quality that’d algo generated playlists create. Like the music is too similar too aligned, it sends my ears to sleep.

I’m dying to build a new music app and it’s something I spend all my time thinking about. I have so many ideas wrt stems and lyrics but I’m personally not willing to take the risk of angering WMG or RIAA.

Would love to have a discussion, email is in my bio.

Would be fun to chat...don't see your email in your bio though.

Ooops, I thought it was. It's in there now.

Reply on news.ycombinator.com