The lie of music discovery algorithms
zeynepevecen.dev
zeynepevecen.dev
Pandora had decent algorithms for recommending things, but it had such a small library that it would frequently repeat the same handful of albums for anything I searched for. This irks me, as I hate wearing out good music.
Spotify is currently where I keep my weeks-long playlists that I've built over the past couple decades. Even with such large playlists as input for their radio recommendations, Spotify doesn't do a very good job recommending new music either.
Whatever happened to the good databases and their algorithms? They definitely used to exist.
Apple has tried to help by using human curated playlists but I frequently find myself thinking I have better taste than their “experts.”
It feels like all the time, when there is human curation factor, it ends up being sold off, gamed, exchanged for favors etc.
It was called a record store employee, and they no longer exist.
There's an AI search engine I'd like to see ...
Even the "market segmentation" of pop music still doesn't work for crap. Even something as basic as "Gee, I like 80s New Wave, how about recommending some artists born roughly in the 21st century who would fit?" seems to be totally beyond the pale of anything currently existing.
If the database quality of current Last.fm were similar to its state back when it had radio, I would think that an AI trained on their data would be pretty good. With the current state of it... It would have to be crap. Heck, even if an AI model could be trained on the play counts of every song of every user on every streaming service, I'm not sure it could approach the curated relational algorithm that Last.fm had at its peak. Would definitely love to see an attempt, though.
I'm happy to not relive the days where CDs were and inflation adjusted $35 for about 10 songs. And there is no economic incentive or guarantee that any given retail employee would know anything about the inventory. Go to home Depot and canvas some of them about home repairs if you don't believe me.
Limited edition vinyl stuff, for example, generally goes for right around that in order to support the artists. Most of that money is going directly to the artist, nowadays (as opposed to live in which it all goes to Ticketmaster).
If you're not willing to spend at least some money, well, then you're part of the problem why artists can't get paid for doing music and why so much of it kinda sucks.
> And there is no economic incentive or guarantee that any given retail employee would know anything about the inventory. Go to home Depot and canvas some of them about home repairs if you don't believe me.
Sure, if you went to Tower Records, you almost always had someone clueless. However, the point of going to those stores was to NOT go to the big retailer, it was to go to the local record stores that had people who worked there specifically because they were super passionate about music.
This was how you found out about that super obscure artist. It was also how you found out about the local bands that might be of interest to someone who liked that super obscure artist. etc.
I've been a daily user of Pandora for something like 10 years. It's been getting steadily worse the whole time, and especially in the last two years.
I like to create my own station that is "seeded" by a few artists, and then allow the algorithm to do what it wants to play related music. This used to be great, until one day I noticed that it had become stuck playing the same 50 or so songs. This was after about three years of listening to that station on a weekly basis. As an experiment, I created a new station and seeded it again with similar artist. Again, it was fine for 2-3 years until it got stuck on a handful of songs. I did this again recently and it has become stuck within four months. There are even a few of my seed songs that it simply ignores and never plays.
Presumably their recommendation algorithm and the Music Genome Project tagging allowed them to, given a set of tags, find similarly tagged songs. This worked really well. It's how they used it and how they picked what track to play next that caused issues.
First, thumbs up. As far as I could tell, at the time Pandora would strongly prefer to play a track you'd thumbsed up on a station over anything else, as long as it was available to play, which mainly just required it to not have played to you in the last two hours. So if you used thumbs up the normal way, a well-used station would eventually turn into a loop of things you'd already heard.
Second, the way they used the algorithm - it seemed to me like they only or mostly used the station seeds as the input, there was no blending going on, and if thumbs ups impacted it, well, they had their own problems. That is, if you had a station with two seeds, it played some songs close to one seed, then it switched to the other seed and played some songs close to the other seed. Skipping would usually bump you to a different seed.
To get around all this and get a variety of new material, I created a station with a lot of seeds - 25 to 50 or more - and never thumbsed up anything on it.
That annoying dark pattern on a piece of software you use? Because there are people who fall to it, clicking on an ad or "engaging" more. That stupid show that keeps being recommended to you? Because a lot of people just sit on the couch, watching something on the list that does not need too much mental processing.
I have a peculiar taste in Music. I love many many different types of music, but once I find a really good piece, I'm not interested in things that are very similar to this one. Looks like if we describe musical work with high dimensional vectors, I like to find good vectors that are not too close to each other. But as the author said, Spotify keeps showing me music that's similar to what I've listened. That's exactly what I don't like (with the occasional exception of something being better than the one I've already found and replacing it).
I assume I belong to a peculiar minority. The recommendation algorithms work very well for predictable majorities.
Maybe someday we have an interesting "musical embeddings" model, and then people can implement personalized discovery algorithms using that?
The more you curate, the more you define your own taste. It’s then easier to describe what you like in a music and triage.
at the end of the day no matter how many times we beat our heads into the same wall, we’re not even close to an accurate discovery model, music nerds are far better at recommendations than any discovery models. far better.
don’t let their insufferability discourage you. you will be too once you start diving into and going on rants about music which is outside of mainstream fluff. it’s like this with any $subject involving wonks. we’re insufferable to anyone who isn’t into our particular genre of technology. food wonks are insufferable, car geeks are insufferable, gamers are insufferable. that’s ok, if you’re looking for someone who is a geek in a topic, you’re likely to become one too :p just be normal around $subject non-wonks and you’ll be fine.
but yeah, music nerds working in a good record store really do know their stuff.
other places:
- music nerd streams on twitch
- music reviewer youtube channels
- college radio stations (most have an online presence) 770 radiok out of minneapolis is incredible
- kexp out of seattle is absolutely amazing (they’re heavily online as well.)
- just about every mid+ sized city has some amazing radio, usually found in the low FM areas.
at the end of the day though, it’s other people. there are far too many variables for every individual which drives why they may or may not like a song at any given moment. other humans are still absolutely unmatched when it comes to navigating this.
Unless your interests are niche. A fun game I used to play as a teen was going with my parents to the record store and seeing if they had any music I listened to online while my parents shopped. Never found a single CD (but they couldn't be that niche, this story is about bands I found out about from my friends at school!). Employees tried to be helpful, but there's only so much they can do when someone comes in and asks for a list of bands they've never heard of.
If your taste is exactly "maximally dissimilar to anything I have liked before," that's actually pretty easy to calculate from the embeddings as well.
That's gotta be how they do it, right? I'm probably wrong.
Maybe I'll build that. Sure would be nice to have.
Nowadays, they have a quite busy research department so I would imagine that recommendation is quite fancy indeed: https://research.atspotify.com
https://www.canburypress.com/products/you-have-not-yet-heard...
First, there is/was no single algorithm, but the core ideas driving a lot of recommendations is:
1. Create user taste vectors
2. Match those vectors to other users or collections of tracks
3. Use that information and combinations of other things to find recommendations.
Each step of the process is constantly being experimented with. Different custom playlists might be using a different combination of tech doing those basic steps.
[Disclosure: Work at Google, but not on that. Just thought that course was particularly well-designed.]
No matter what I do in Spotify, under several different rounds of accounts, it always seems to gravitate towards the tastes of the general public, i.e. some form of mass-market pop.
Their recent "ai" assistant was a slight improvement because you can ask it for less popular music which is typically better for music discovery.
Maybe these music services should ask you for music you hate, and start from there instead.
They have a huge sample set of stuff I like just from that (it's still accessibly via YouTube Music, but it sucks). Still, any time I fire up YouTube Music and play a song, the next 2 or 3 can be solidly appropriate and often stuff I wouldn't have thought to play - in a good way.
Then there's a sharp decline where it just starts playing the same thing it played last time or the most common song off the most popular album from the most well known artist adjacent to what played previously.
The whole reason I use the "algorithmic feed" is because it should be optimized for my tastes and what will keep me listening. I still "thumbs-up/down" stuff but it never seems to work. You would think with their insane library and huge dataset of my prefs and listening data they'd be able to generate something great.
Instead I just go back to the human programmed Shoutcast stations I have bookmarked after a few songs.
It's the classic problem of "Ok, but how do I sort these by which ones are actually any good?!"
Are you sure this is an accurate description of your taste? Or do you mean "I'm not interested in things that are similar, but lower quality"?
I'm very much a weird-music enjoyer, and I often have the latter problem where "similar" songs actually just don't capture the same vibe as the truly engaging new song I just heard. But that's not because the algorithm is choosing music that's too similar; it's the opposite. It's trying to choose something similar but can't, so it just picks the next-best thing which I actually don't like.
In my experience, and the reason I left Tidal several years ago, is that they lean heavily into modern hip-hop, compromising relevancy for the sake of promoting "friends of the company" (Jay-Z, Beyonce, etc).
More importantly though, Spotify doesn't directly link artists to bands they're in. For example, on Spotify, Billie Joe Armstrong is credited with his work on Green Day, but if you go to the Billie Joe Armstrong page, he isn't. On Tidal, each artist can also be broken down by their role on a project. For example, you can see just the songs that Paul McCartney wrote (turns out he wrote a track for Drake called Champagne Poetry in 2021... who knew). On Paul McCartney's page on Spotify he isn't even attributed to The Beatles
Unexpected writing credits like this usually indicate sampling. In this case [0], Drake sampled a song [1] that sampled another song [2] that was a cover of a Beatles song [3].
0: https://en.wikipedia.org/wiki/Champagne_Poetry#Samples_and_c...
1: "Navajo" by Masego
2: "Michelle" by The Singers Unlimited’
3: "Michelle" by The Beatles
Hey Jude by The Beatles - Tidal lists 22 different people who worked on this song. Spotify lists 3.
Last.fm was good, but inevitably went downhill when it was bought out. As they had to commercialise more and more from the early days as a uni project that became audioscrobbler, with a tiny userbase, they followed the standard Doctorow model of capitalist decline. They had a sweet spot a few years in when they had plenty of data flowing and they hadn't yet messed up their APIs. Right now I use it to log my listening but I'm waiting for the email that says I'll need to move my data elsewhere.
You can set the seed artists, then select levels for "artist variety" and "music discovery." You can also filter based on tags.
"They are not suggesting new, very interesting melodies. They are finding you the tweaked versions of the songs you already like and, even on your first listen you can predict the melody that’s to come."
(If you play music for your toddler, your future suggestions will include children's music not because it sounds similar but because "a critical mass of other people who listened to Baby Shark on repeat also listened to: Old Town Road")
It is weird and it’s ironic that they call that “discovery”, as it feels more like variations of what I'm already listening to.
(not trying to defend those suggestion algorithms, just analyzing the comment)
To add to this analysis, I think there may also be a feedback component to this problem that exacerbates the issue, since most users are passively using the suggestion algorithm.
In other words, if the suggestion algorithm tends to create a homogenized persona of the user's taste, say, because they don't bother to actively correct it, then this persona is embedded into a cluster of people with similar personas. And because the persona is now closer to said cluster, the suggestions will become even more homogenized. Moreover, since the cluster is mostly composed of passive users, the cluster itself will tend shrink (eg in variance) and to get more homogeneous.
I suspect that most algorithms do not do enough to prevent this global trapping effect, and so even if they have some method to sample "something new" for the user this becomes less and less efficient as more users rely on the algorithm for their suggestions.
It's easy to see how this would've been baked into a human-made algorithm when you consider waveforms. Speaking only to Spotify's algorithm here. And it doesn't really bother me for obvious reasons. But it is creating something of a musical echo chamber for me.
I don't understand why is it so hard to offer something along these lines:
1. Define dominant user preferences by clustering and segmenting the field of listened genres.
2. Build a list of relevant "neighbours":
2.1 Manually added users/friends
2.2 For each of the dominant genre preferences, find users with a high level of artist intersection within that genre and add them
3. Now, for a "find similar" query:
3.1 Define a reasonable time window
3.2 For each neighbour, find points in time when they listened to the queried track/artist
3.3 Build a list of tracks/artist from the defined window around the points found
3.4 Filter tracks/artists that are too "distant" on the general genre/tag map, or lie outside of the user's dominant preferences (with a degree of boundary feathering, perhaps)
3.5 Filter if similar to negative part of the query
3.6 If novelty is required: filter artists/tracks according to the degree of their presence in the user's historyThe cost to play a song is expensive, so if you can actually profit by putting a new artist instead of paying, why wouldn't you?
Sure your customer gets 3 minutes of potential garbage, but they don't realize that they generated revenue for the company just by sitting through that song.
If you give your customers a great experience, they are going to listen to more music, which is bad for the bottom line.
There doesnt seem to be any competition due to IP laws, so there is no incentive to be good.
1. The curse of dimensionality when computong distance functions and
2. The cost function of a bad song. People get mad if they get too many things they don't like. Notice Pandora immediately sends you back to the band that seeded a station upon any thunbs down.
"We have [favorite band] at home" - it picks things you like from your favorite band - instruments, tempo, etc then finds bad knockoffs that are superficially similar but painful to listen to.
The "Iron and Wine" problem - some bands are so generic that they tick every single similar box and flood your recommendations. For years, it didn't matter what band/genre I tried to find recommendations from, I got Iron and Wine.
If I want music "like" "Groove is in the Heart", is it because:
* I want mid-tempo house-like dance music
* I want major key songs with female singing
* I want songs with rap interludes
* I want 90s music
* I want fun party music
* I want music that reminds of that awesome trip I took with my friends a few years ago where we played a bunch of songs over and over
There is no right answer to this question. But, outside of just looking for playlists, no music app I've seen gives you a way to specify in what way recommended music should similar to the current song.
I see this effect most acutely when I listen to something that happens to be popular. For many people "heard it a lot when doing this fun social thing" is one of the main reasons they like a particular song. This was true for me too when I was younger. But for me today, I'm mostly oblivious to popularity. I just like stuff that sounds a certain way.
Whenever I stumble onto a song that has a particular sound I like that happens to be well-known, the recommendation algorithm just starts throwing other popular stuff at me that sounds totally different.
* Musicians having fun with instruments.
This is pretty bad if you have strong feelings about how much screaming a metal song should have. There are songs that fit exactly what I like except for that variable and Spotify does not get that I keep skipping them for a reason. It's rarely a "bad knockoff", but it definitely hits "painful to listen to".
It's really strange to me since it successfully creates playlists around different types of music that are sort of similar but shouldn't cluster together.
I have noticed an interesting phenomenon around TOOL. If you start a playlist on Apple Music from TOOL it will start playing everything from Metallica to Nirvana. A lot of people like TOOL for a million different reasons and Apple doesn't know any different except for the overlaps in taste. If you play a Mike Patton band, such as Mr. Bungle though -- you will get some TOOL in your playlist -- because both bands are esoteric and often challenging.
I'm looking forward to the day (or wishing maybe) when my app considers these factors. For me the issue isn't discovery, but rather I want my robot DJ to vibe more closely with me.
That tends to disregard many reasons you like a particular track, and does especially badly when the liked-track isn't part of a uniform style for an album or artist.
I recognize it's a heck of a lot easier to implement, but it's still a disappointment.
https://developer.spotify.com/documentation/web-api/referenc...
The guy behind Every Noise at Once (engineer at EchoNest/Spotify until the recent layoffs), has some interesting thoughts about this topic:
https://www.furia.com/page.cgi?type=log&id=478
He’s quite biased towards not using ML or acoustic characteristics for recommendations. But even if you disagree it is interesting to hear about how things were working under the curtain (for daylist in this case).
I have no idea if anyone is listening to them, though, because there doesn't seem to be any feedback system for the community playlists. That would be a useful addition, IMO. If TIDAL doesn't want to pay dedicated staff to curate playlists, they could at least make some way for the member-created playlists to get featured or gain reputation.
My solution: listening to NTS, an eclectic online radio station, where diverse artists create playlists.
[1] Technically "random song not in listen history" would work out, if you'd really like to call that a recommendation algorithm.
The NTS app is great: for Web, Android, iOS - it's always being steadily improved. A very nice feature to aid discovery/curation is that every track in a tracklist has a 'copy song and artist info' so you can easily search for tracks on your streaming platform. Not sure if this is a subscriber only feature.
I also use the 'identify song' feature in the Google search app on my phone, similar to Shazam.
If the algorithms aren't doing it for you then do yourself a favor and head to https://nts.live
Nothing beats that one friend who used to DJ and still obsessively digs crates.
Radio Paradise is very much a rock station at heart, so necessarily for everyone's liking. If you're into classic rock mixed with contemporary rock, mixed with a bit of everything else, it's worth a shot.
that dj friend or like i said in a different comment, your local record store employees.
college radio stations.
and just other people. it really is that simple.
People who feel they have a calling to be a DJ sometimes actually do.
I have a reasonable list of TuneIn stations (mostly US) that provide my favorite “discovery”.
All music recommendation engines at this time still aspire to be mediocre, they aren’t even playing the S.A.,e game as a human who is good at it.
Unfortunately, such humans are unevenly distributed.
However, even in the various artist stations, they do play artists "similar to" as well. Which is how I started listening to stuff like Murder By Death, Eilen Jewell, Bakar, Black Pumas, and others.
> I come up with an idea of generating playlists from images. Images that you shot on your phone yourself, or that you found on Pinterest, or a painting that you really like and feel inspired by.
This is genuinely interesting! Do you send the images to an LLM with a prompt like "generate a list of songs that would go well with this"?
Yes, I have a prompt like that the current prompt is this:
'You match the vibes of the pictures with the right songs and turn them into a 3 song playlist with a playlist name. The music genre of the playlist should be consistent for each song. Be creative with music selections, explore different music, be consistent in terms of the genre of the 3 songs. The playlists should be provided in an object array format, like this: \'[{playlistName: "string", songs: [{songName: "string", artist: "string"}, {songName: "string", artist: "string"}, {songName: "string", artist: "string"}]}]\'. Do not add any other text information and only give outputs in the provided format. Your playlists must match the visual vibes and maintain the specified format without any additional information.',
YouTube of course doesn’t care because they don’t make more money when you see something awesome.
That makes things worse IMO. My favorite videos tend to be one-offs not channels producing regular content. Unsubscribing from everything definitely improved my feed.
I think YouTube's recommendations used to be excellent, especially for music, but I've personally found it to be terrible recently. It no longer recommends anything new to me, and I suspect that it's way over-tuned. If I see a video that looks mildly interesting I'm a bit hesitant to watch it, because I don't want YT to decide that it should become 50% of my feed for the next week. Which from the recommendation system's perspective is just weakening the signal I'm feeding to it even further.
I also feel the same as you regarding the youtube algorithm. I actually get better recommendations sometimes by just logging out since it will try showing me new stuff.
One thing I'm not sure about is whether it's actually the algorithm's fault or if my expectations have become unrealistic and made me lazy. I used to read magazines and blogs to find new music. There are still tons of people writing about their favorite music, labels that act as curators, etc. I just don't seek them out and instead expect to be spoonfed by the algorithms. Even if this is true though, I suspect many of these algorithms could do a better job.
Also RIP Netflix's old recommendation system. I guess it wouldn't make sense when they can't license every movie like they used to, but I remember it being great. Although maybe it was just pretty good and I was younger and less familiar with the back catalog of good films.
That's the other thing I wonder - am I just getting older and less excited about new things? There used to be a real vitality to finding something new and exciting. Now it kind of feels hard for anything to feel that fresh anymore, it all seems like variations on the same core ideas. I do still find new stuff that I like, but it doesn't have the same thrill. Maybe I'll always be chasing that dragon of youth haha.
Most cities and metro areas with more than a few hundred thousand people have jazz, punk, indie, hip hop, country, choral, and classical scenes. Certainly true of any ville with a university.
Check out a local weekly, listen to college radio, look at the online calendars of local venues and clubs, take a risk, check something out you've never heard of before. You may be surprised. There are musicians and scenes which fly under the radar of widespread Spotify and Youtube popularity which nevertheless deliver great performances and often themselves lead to other new, interesting discoveries.
A side effect is you may also end up talking to someone at these gathering places and making new acquaintances: again, another great way to discover music and other things.
There is so much that has been built online whose subtle or sometimes overt goal seems to be to eliminate actual human contact. I suppose that is attractive for some, but I feel the opposite is what a lot of people yearn for, and it can be achieved with just a little bit of investment.
Maybe in Asia it's different, but in Europe and in the USA only 10% of the people live in a city with more than 300k inhabitants.
Many of you were also interested in knowing what's going on behind the scenes. The setup is pretty basic. I've built a NextJS app, for the LLM model I am using open AI gpt-4-turbo and sending the images there directly without any database for images. I did a little prompting to get the same output everytime and when I get the output I make search on the Spotify API, find the songs and create the playlist with them on your own authenticated spotify account.
Likewise I also don't have a database for the emails eighter. I am using spotifys authentication.
As I said before this was just for me at first but it is very exticing to see many people interested by this idea.
If you don't trust openAI models and don't wanna send any pic data to them please don't write your email
So what worked for me in the past is finding less popular artists and then checking their similar artists.
This seems like the complaint of somebody who hasn't been using spotify very long. After a decade plus, I feel like my algorithm is a rich compost pile of all of my previous phases of music. Spotify is excellent at letting me broaden my horizons or jump down a rabbit hole from a random starting point, like a song I hear in a public space or commercial or something sent by a friend. Maybe the OP should keep their ears open to more sources of randomness from the outside world?
It probably helps that the strongest areas of my taste are relatively small or niche genres, like Scottish trad and Celtic (folk) rock. In those niches, similar-but-different is often distinctively different in actual experience. Sure, there's covers of the same song from time to time, but I actually do like enough of those not to be bothered, if they bring something new.
My solution (doesn't work for everyone): I have a large library on the microSD card on my phone, and set the music player to Shuffle. Quite often a song comes on and I think, "Wow, I own THAT??"
OK, I'll admit that doesn't play any new music. However, no bills for bandwidth!
Why? It's almost exactly the same experience you get from all the competing services, and that experience is fantastic: any album, any song, instantly ready to play.
> I have a large library on the microSD card on my phone, and set the music player to Shuffle.
This is a fine situation to be in but from the perspective of your fellow music consumers it is highly undesirable. Music libraries involve a lot of time and money. For the cost of one album per month you can subscribe to an unlimited service.
> no bills for bandwidth!
This definitely sounds unique to your geographical situation.
I admit that streaming services are bad for artists who want to make money from album sales, but asking consumers to use something else would be like asking people to ride horses to work in order to keep farriers in business. The 1990s are long gone; being a famous musician isn't guaranteed to make you rich anymore. Meanwhile the not-famous musicians who make up 99.999% of the music population can happily continue not making any money off of album sales the same way they've always done.
You got that right.
This is a totally vapid, progress-is-great response. Do you work for Spotify?
Read some Ted Gioia and get cured of that.
The premise that many folks miss here is the idea that Spotify is, at best, thinly interested in recommending music that is good for YOUR interests. Spotify is the music business, and specifically the pop music business, has long discovered that's it's much more economically expedient to force feed musical taste onto the public than it is to chase the whims of organic hit-making. Payola is as old as recorded music. Spotify recommends what Spotify wants it's users to listen to. They have all kinds of side deals and marketing deals with labels, they have cheaper costs/royalties on some tracks than others. Popular tracks cached in their CDNs are probably cheaper to recommend than long tail ones etc. They have strategic priorities like gaining on apple for podcasts, and therefore injecting allsorts of podcast recos in the UI wether you asked for that or not.
Music discovery used to work for me with pandora, nothing else has. I have no idea if they had better algorithms, or just a better catalog. It doesn't seem to work as well as it once did either.
Tl;dr I don't think it is simply your preference for lyrical content over melodic content that causes the algorithms to fail. They are just bad.
I feel like there are two kinds of singers - people who are good at singing, and people who have something to sing about. You and I, I think, prefer the latter.
> Marvin Gaye, Public Enemy, Rage Against The Machine, Beyonce
Thanks for the laugh, man, I needed it.
Plus, there is no real explanation on the page as to how this would work, not even from a high level.
Likewise I also don't have a database for the emails eighter. I am using spotifys authentication
Next to that there is also the shadow fact that there are millions of dollars and more in fake plays, subscribers , ... SMM panels are getting big because the music industry doesn't allow for anything less than instant fame.
There are smaller services yes, which allow for independent promotion and distribution (Last.fm, RateYourMusic) but these have fairly obvious flaws in how the listener can approach new music (RYM pushes ratings first and foremost, and both last.fm and rym push trending artists to their users).
Instead, because the value of music is zero (really, negative since the number of listens, streams, album purchases, etc can fail to recuperate the cost to make it ), the act of distributing music presents economic risk unless the release itself can be controlled by the investors through advertisement or paid promotion.
First, it's hard, as we all know.
Second, it turns out that the "best $ITEM discovery algorithm" accolade doesn't pull in that much extra revenue. It's far better (for them) to use people's expectations of such an algorithm to profit from a bait-and-switch. As an example, see Amazon's search engine.
If I'm in the mood for a certain mood/genre, that's what I want to listen to. Even if a song comes on that I normally love, if it doesn't fit the mood, I want a way to say "that doesn't belong here right now". So in Pandora I start a station and can thumb down songs that don't fit that station's theme, and Pandora understands that doesn't mean I don't like that song. It just means I don't like that song in this context.
As far as I've seen, every other service only registers these globally. Either I like a song or I don't. That doesn't make any sense to me.
It's all rather random but relies somewhat on your gut instinct. I find it more enjoyable than the top music streaming services. Case in point: someone uploaded an excellent field recording of a Bruce Hornsby concert from 2017 yesterday - listened to the whole thing a few times already (and I'm not really a big fan, but he's a great showman).
If you like old-school metal, here's some great youtube channels (no affiliation to myself):
https://www.youtube.com/channel/UCCGbKiCJjph8Grazqmo7z4w https://www.youtube.com/channel/UCD5Ny_jQ8cs9JXVPWXg9iNw
Support these small bands.
I find their comprehensive section of Lists (featuring lists from all the major publications) and aggregate lists is inspiring for discovery.
For example, check out this list of the best albums so far this year according to The Quietus, containing some great stuff your algorithm would never consider:
https://www.albumoftheyear.org/list/2284-the-quietus-albums-...
I always thought doing that with the entire spotify library would be amazing. Give me a low dimensional space to explore the library. Even cooler if the embeddings have similar geometric properties of language embeddings where I could do arithmetic with songs to find interesting combinations.
Back in the day, this was friends lending me CDs. Now, I follow a bunch of people on Mastodon who almost only post about music.
I've also found following the releases from specific smaller record labels quite useful - often the artists on a label will fit a certain vibe, even if they don't specifically overlap in genre.
* There was a thing called "Library" in addition to "Likes". Basically all "your" songs, not necessarily liked you.
* When clicking on a "Feeling Lucky" button, it selected a random song from Library, and started an auto-generated Radio off it.
It allowed to listen to random songs based on your library. I miss that in Spotify.
However I would prefer a service that allowed me to tell what I don't like and then use that preference to filter out everything similar to it.
this is clearly no longer the case for any major streaming platform. their own logic to promote might be too egregious now. same seems like for shuffling through a large playlist.
one can try to empathize to the ones designing this (e.g. shuffle anticipating network drops and switch to cached results for the next track) but self-discovery will remain evergreen.
graph traversal playlists are the most interesting idea to me, especially if you can put some bounds on (i.e. weight positively and negatively certain artists in the graph)
Please get in touch
It's noted that her current tour is sponsored by Spotify.
Otherwise, most of the services (Spotify, Apple Music, Youtube Music) have been really bad at recommending music. No amount of downvotes on songs seems to give their algos any input on bands I don't want to hear. Further, they always seem to devolve in to ridiculous recommendations. I've asked for "Louis Prima" radio and have them insert hip-hop or rap.
Google Play Music did better than those 3 but that died.
Likewise I also don't have a database for the emails eighter. I am using spotifys authentication
If being permanently locked into a single intelligence service (or risking getting cut off or sued) is unacceptable for you as for me then OpenAI terms today are not acceptable. Yeah yeah maybe they won’t go after you, but why miss the opportunity for malicious compliance? Ditto Claude. Try a specialized model for your use case instead.
Gemini has no such customer noncompete, and with llama 3.1 meta removed theirs last week.
What do you mean, "app"? There's no "app" there.
If anyone constructed a PDF, which was itself blank but, via embedded JavaScript, loaded parts of itself from a remote server, people would rightly balk and wonder what on earth the creator of this PDF was thinking — yet this is precisely the design of many “websites” [1]
Weirdo.