Spotify - anyone heard of it?

^^^^^^

finally got a really good Discover Weekly playlist this week

discover weekly is uneven as hell, last two weeks it was like "why do u think i like any of this lol spotify, ur dumb" and then the one i got for this week is pretty damn all right thus far

It certainly seems determined to try and get me into the whiny voiced indie that I haven't listened to for 15 years. Still, about five tracks per week pique my interest, maybe not a great percentage but probably no worse than back when radio was my regular discovery channel.

is this all because it's using your playing of last week's recommended racks in figuring out this week's?

(is it a spotify home-grown thing or does it use a recommendation engine like gracenote / echonest / senzari?)

my weekly discover playlist seems better, too. i am very susceptible to the enthusiasm of others, though.

My Week 1 Discover playlist was excellent, but Weeks 2-4 have been way off the mark. However, all my partner's playlists have been bloody marvellous. He uses Spotify much less than I do, and only to listen to music that he already loves - whereas I'm forever dipping into new music speculatively, with inevitably mixed results.

ugh i still don't have discover weekly yet, thanks obama spotify

My Week 1 Discover playlist was excellent, but Weeks 2-4 have been way off the mark. However, all my partner's playlists have been bloody marvellous. He uses Spotify much less than I do, and only to listen to music that he already loves - whereas I'm forever dipping into new music speculatively, with inevitably mixed results.

― mike t-diva, Monday, August 10, 2015 2:11 PM (43 minutes ago) Bookmark Flag Post Permalink

this has been my experience -- week 1 was good, subsequent weeks were "but I already *have* opinions on banks and bat for lashes and the knife," this week was either completely off the mark or (more likely) I was for however many weeks

For a while I was getting a lot of tracks I had already listened to on Spotify, which was odd. Just looked again now and this week's doesn't have that problem but mostly recommends me things from the most popular/indie end of the spectrum I listen to, i.e. things I've already heard of and decided I don't care about checking out.

Of course ilxors are a really hard section of the population to recommend music to!

i think the thing about recommendations for me is, i generally don't care about music unless it comes with a story, or a connection to someone i know. it's a way of changing yourself via another's love. like trying on a shirt i'd never imagine wearing but finding a way into it via the fact it was a gift from somebody special. the music "itself" without those conmections is just kinda... flat to me. usually. that's why i like music writing, and why i like friends who are passionate about music. radio can act as a sort of placebo via a DJ. anyway when spotify, or last.fm, or whoever, recommends me something it's just dry and it could be the greatest thing ever but is hard for me to latch onto.

yes, but if a recommendation gives you a song you've heard with friends but didn't know what it was at the time, that system partially serves the need. i see your point though.

i'm a massive introvert so i've always discovered music the hard way. these 'curated' services suit me pretty well, in that i no longer have to get off my arse and look for new music that's similar to what i like.

and, my friends and forums and fan sites all still exist, so it's just an additional way to find new stuff.

Spotify has an equalizer in the settings on iPhone!? how long's that been there? i thought they'd have shouted it from the rooftops.

What?!

yeah a few releases back! waiting patiently for the desktop version

we should start a Recommend Tracer Hand Some Music thread

the rest of this site works well for that but a personalized thread would be fun. i would review everything!

Anyone else following this spotify playlist Fresh Finds

https://open.spotify.com/user/spotify/playlist/3rgsDhGHZxZ9sB9DQWQfuf

updated every wednesday

non-personalised new music from unknown / emerging artists

i tend to skip through it, i.e anything that starts like dullard paste style 90s American indie rock - straight to next track. however it's worth going through for a few gems

this is my second week experiencing Fresh Finds - Meltybrains? a band from Ireland that i have been following for over a year are on this weeks playlist with their afrobeat-futuristic-electronic-art-rock-pop track Vine

take a listen:

Meltybrains? - vine
https://open.spotify.com/track/3JTW12iyImNivo7O6jSqsb

INSIDE SPOTIFY'S PLAN TO TAKE ON APPLE MUSIC
http://www.fastcompany.com/3049231/tech-forecast/inside-spotifys-plan-to-take-on-apple-music

It's called Fresh Finds. At first glance, it's just another playlist—it could easily be one of the collections hand-curated by Spotify's 32 music editors, or one of its 75 million users for that matter. But Fresh Finds is different: The weekly playlist is generated using a chunk of the predictive big data technology that Whitman and his team at The Echo Nest brought with them to Spotify last March. The mission of Fresh Finds is to identify under-the-radar artists that are generating buzz online and surface the ones most likely to break out.

"Fresh Finds is a distillation of the hippest users on Spotify," says Whitman, pulling up a list of 38 tracks projected against the conference room wall. "These are the artists that are going to break out soon because they're being listened to by these people."...

How does he know? The machines told him, naturally. Fresh Finds takes a central component of The Echo Nest's original methodology—its web content crawler and natural language processing technology—to mine music blogs and reviews from sites like Pitchfork and NME and figure out which artists are starting to generate buzz, but don't yet have the listenership to show for it. Using natural language processing, the system analyzes the text of these editorial sources to try and understand the sentiment around new artists. For instance, a blogger might write that a band's "new EP blends an early '90s throwback grunge sound with mid-'80s-style synthesizers and production—and it's the best thing to come out of Detroit in years." If this imaginary act goes on tour and writers in Brooklyn dole out praise of their own, the bots will pick up on it. It helps address an issue some people have voiced early on with Apple Music, that its selections aren't adventurous and it tends to recommend things you already like rather than things you might like.

After generating a list of talked-about artists, the Fresh Finds algorithm checks that list against the listening habits of a subset of Spotify's users that the system has determined are "ahead of the pack"—that is, the many thousands of very active listeners who tend to discover new artists before they break. In early, internal testing, Whitman and his team say that this approach, in addition to being clever, actually works. He shows me a collection of analytics charts that compare an artist's "buzz" factor (as determined by the blog-reading bots) with its listenership on Spotify. Sure enough, in many cases, a burst of online chatter about a new artist is followed by a spike in listening.

read the whole article..for all the details

I worry that if I listen to Fresh Finds, it might give me another shitty Discover playlist.

Wary of it due to the usual problem with centralised trend aggregating as described.

Highlighting curated playlists of new stuff (NOT divided by genre) from a range of individuals is preferable.

listening to a track once will not ruin your discover playlist, however if you only save the tracks you do like to another playlist then your discover playlist will likely improve

as explained further in the article

http://www.fastcompany.com/3049231/tech-forecast/inside-spotifys-plan-to-take-on-apple-music

Discover Weekly analyzes your own listening history, with an emphasis on what you've been bingeing on recently. It then takes that knowledge and compares it to the playlisting behavior of other users in the hopes of tapping into that distinctly human, gut feeling that dictates why one song sounds good following another or why this particular collection of 35 songs feels perfect for your mood on a given afternoon. Scanning millions of playlists, the system tries to find tracks that are commonly listed alongside the music with which you're already familiar, and then group those tracks together into a new, personalized playlist. Effectively, it takes the classic "people who like that also like this" logic of collaborative filtering and applies it to the process of making a mixtape for somebody.

This is kind of an obvious thing to do, but it's been entertaining me recently:

From the main everynoise.com map, go to the page for any genre and click on "List" at the top of the page. You'll get an embedded Spotify playlist of songs matching that genre. By clicking the </> icon in the embedded player's controller, you can copy the playlist link, open it in the web player or stand-alone app, and then follow that playlist. This is convenient for me because I usually listen to Spotify on my home sound systems or iPhone rather than on my computer.

Here's a playlist for "cool jazz" that seems representative: http://open.spotify.com/user/thesoundsofspotify/playlist/3RtFvzIXD7ulUCXkWdIOWW

How We Understand Music Genres is a good brief explanation, but I'd love to read a more detailed discussion of the data analysis behind Every Noise at Once.

More on Fresh Finds

Understanding the brand new
http://notes.variogr.am/post/125515460365/fresh-finds

by Brian Whitman

Scientist, sound artist, co-founder & CTO of the Echo Nest, Principal Scientist at Spotify.

“Fresh Finds” is made possible by a scalable analysis of the musical activity happening outside Spotify: daily, we automatically find artists people are talking about on music blogs and news sites more often and with more intensity than their playcounts should suggest. These are the artists that find fans through word-of-mouth, shows and the hard work of making unique music that connects to at least one person. We then filter those artists through a real time analysis of Spotify listening behavior and weekly generate a list of brand new songs that we think we will gain in popularity the next week.

The listener activity that happens to music deep in this brand-new, unheard part of the spectrum is hard to automatically understand. It comes from nowhere, and people discover it from other people, often outside of our platform. They read music blogs and press, or a friend in the know passes on a link. It’s not based on popularity or audio or likes or clicks. Some of these Fresh Finds had virtually no plays when data indicated we should publish them on the playlist. Watching a brand new artist release a brand new track with no connection or external push and seeing it at first slowly, then rapidly gain in plays on Spotify has been life-affirming.
I see recommendation, filtering, or prediction of this “brand new” as a new artistic frontier in music understanding. Every music data scientist wants to help artists and listeners, but the quantity of known good that a precise recommendation for a well known band earns all but vanishes when stacked up against the connection between a new unknown and her new fan base. I’m ecstatic to help that process even a little bit.

Brian Whitman's Favorite jams from the past 6 months of Fresh Finds:

Found on Fresh Finds playlist:
https://open.spotify.com/user/bwhitman/playlist/1EViHawT7ibM3pS2gPE4mW

Since Fresh Finds started working internally for us early this year, I’ve discovered more music there than from any other technology or web site or service I’ve ever used.

There are direct links to those The Sound Of playlists at the bottom of each genre page. Save you a few clicks...

Anyone else following this spotify playlist Fresh Finds

https://open.spotify.com/user/spotify/playlist/3rgsDhGHZxZ9sB9DQWQfuf

updated every wednesday

non-personalised new music from unknown / emerging artists

brilliant discovery! it's a proper lucky dip, and i'm enjoying at least half of this week's. fun stuff like this will keep me on spotify indefinitely.

There are direct links to those The Sound Of playlists at the bottom of each genre page. Save you a few clicks...

Aha, thanks.

So digging into Fresh Finds right now ... I like the idea.

Tracer Hands' post about music recommendation is, er, spot on for me as well. One thing I wonder is if music recommendation engines could do a better job going forward of unpacking WHY a song is recommended to you in some way -- beyond "you listened to X, you should try Y" which is a pretty shitty corollary. Like TH, I tend to be much more excited about a recommendation when I have why it was recommended to me explained a little. For something like Fresh Finds, it would be interesting if you could essentially see the blogs/reviews/Soundcloud momentum that, even algorithmically were responsible for putting into the list. Is that possible within the Spotify mobile app? Not sure. But if Spotify ever let other companies license their stream to build apps that house it, my sense is that some enterprising technology company could find a way to demonstrate this in quasi-infographic form, with little blurbs, factoids and so forth. The current of just seeing a list that says "Trust us -- we have some scientists and dudes with impressive beards" isn't really as powerful.

To put this another way ...

Right now we're still sort of at the “Hey you like Kraftwerk, have you heard Devo?” side of things. Fresh Finds and Discover Weekly sound like they're operating more at the "You like the sparse little classical melodies and dislocation in Kraftwerk’s music -- there’s this other band that kind of does that that’s getting a ton of buzz on Pitchfork and Soundcloud you should check out” phase.

It would be cool to see the buzz factor brought to life somehow.

What would be great is a website where people could talk about loving music, maybe post about the reasons they love specific music or recommend things to you based on your own past verbal characterizations of your tastes and preferences. Not sure how you'd measure what music they like most... some kind of voting-based mechanism would be good but I suspect only obsessed weirdos would actually put in the time. Ideally it would have a really minimal, low-impact interface (maybe even just text on a white background?) for those of us using it on older devices. I'm not a technical guy though and it seems like that kinda thing might be beyond the technical capabilities of Spotify. Advertisers might not go for it either, I dunno.

would never work

Yeah, most people would tire over discussing music and instead talk about their love of everything else.

you guys

Steve Hoffman forums

Glenn - point EN's computer at ILM. ;)

Actually, I'm sure you could get great recommendations/correlations from the ILM archives. Sell the datamining rights off to Spotify and ILX could be funded for decades.

Then everyone will complain that Paris Hilton or Big N Rich are being recommended to them

New data reinforcing that old T Swift pain

hey guys my inside algorithmic sources indicate that this band 'radiohead' is going to be big big big - don't know about you but i'm getting in on the ground floor

Listening to ILX Listen - 2015 Spotify Genre Playlists
still waiting on my check tbh

(some of those playlists have 90 subscribers now btw)

One thing I wonder is if music recommendation engines could do a better job going forward of unpacking WHY a song is recommended to you in some way -- beyond "you listened to X, you should try Y" which is a pretty shitty corollary.

pandora used to (maybe still does but i haven't used it in a long time) have a little note on each song saying why they decided to play it based on the parameters of your song/album/artist playlist; "we chose this song for you because it's a power ballad in a minor key" or something. it doesn't seem like too much work to add some more factual info on the artists, and maybe why they're trending, to pop up as well.

re: Glenn - point EN's computer at ILM. ;)

forksclovetofu is OTM, i was just about to mention this !

note, ILM data for 2015 music is ALREADY in the spotify data mining sets i.e the ILM Rolling spotify playlists that forksclovetofu collates. Those playlists will be compared to similar playlists that will influence the discover weekly playlists for many other spotify listeners. So if your already listening to the ILM spotify playlists regularly your benefiting from the existing collective intelligence of ilm and other likeminded / similar curated playlists.

going back to Fresh Finds,

I found this

The Firehose
https://open.spotify.com/user/predict0/playlist/3mZgwuBZqoRWIQqQB7IsTb
Fresh Finds Pro Edition

only 24 followers of this hidden away playlist. Probably most of these are internal spotify employees, but anyone can follow.

over 400 tracks in the last 4 weeks - tracks added regularly - 32 tracks were added thursday morning.

so, this is like a spotify rolling recent version of the Hype Machine for new tracks for new / emerging artists

hype machine
http://hypem.com/

there is also this playlist:

Blogmergence
https://open.spotify.com/user/predict0/playlist/0L5QOqPKhNaYOx1IlUdjH6

this looks like a unfiltered data dump of recent tracks crawled from music blogs. 47 tracks were added on thursday morning.

this is the spotify R&D account:

predict0
https://open.spotify.com/user/predict0
useful to follow

as refereed by Brian Whitman, in the link i mentioned upthread

http://notes.variogr.am/post/125515460365/fresh-finds

they have created human filtered best finds for july / june / may

regarding what spotify need to do next, they need a proper blogging / content strategy outside of the spotify platform. i.e to start a new music discovery blog. They have content curators i.e playlist makers & data scientists but spotify are underutilizing their knowledge and enthusiasm.

http://www.nme.com/news/various-artists/87588

President Obama uploads his own personal playlists on Spotify

thanks obama

lol reminds me of this - https://mobile.twitter.com/blippoblappo/status/460194112142979075

surprised he's so into death grips, but hey

tbrr finding out that michelle and barack's getting it on mix is fairly close to my own is kinda awesome

maybe you can use that along with the knowledge of his murders to delay orgasm /morbs