Posts Tagged Music
Yesterday, SXSW opened up the 2012 Panel Picker allowing you to vote up (or down) your favorite panels. The SXSW organizers will use the voting info to help whittle the nearly 3,600 proposals down to 500. I took a tour through the list of music related panel proposals and selected a few that I think are worth voting for. Talks in green are on my “can’t miss this talk” list. Note that I work with or have collaborated with many of the speakers on my list, so my list can not be construed as objective in any way.
There are many recurring themes. Turnatable.fm is everywhere. Everyone wants to talk about the role of the curator in this new world of algorithmic music recommendations. And Spotify is not to be found anywhere!
I’ve broken my list down into a few categories:
Social Music – there must be a twenty panels related to social music. (Eleven(!) have something to do with Turntable.fm) My favorites are:
- Social Music Strategies: Viral & the Power of Free – with folks from MOG, Turntable, Sirius XM, Facebook and Fred Wilson. I’m not a big fan of big panels (by the time you get done with the introductions, it is time for Q&A), but this panel seems stacked with people with an interesting perspective on the social music scene. I’m particularly interested in hearing the different perspectives from Turntable vs. Sirius XM.
- Can Social Music Save the Music Industry? – Rdio, Turntable, Gartner, Rootmusic, Songkick – Another good lineup of speakers (Turntable.fm is everywhere at SXSW this year) exploring social music. Curiously, there’s no Spotify here (or as far as I can tell on any talks at SXSW).
- Turntable.fm the Future of Music is Social – Turntable.fm – This is the turntable.fm story.
- Reinventing Tribal Music in the land of Earbuds – AT&T – this talk explores how music consumption changes with new social services and the technical/sociological issues that arise when people are once again free to choose and listen to music together.
Man vs. Machine – what is the role of the human curator in this age of algorithmic recommendation and music. Curiously, there are at least 5 panel proposals on this topic.
- Music Discovery:Man Vs. Machine – MOG, KCRW, Turntable.fm, Heather Browne
- Music/Radio Content: Tastemakers vs. Automation – Slacker
- Editor vs. Algorithm in the Music Discovery Space – SPIN, Hype Machine, Echo Nest, 7Digital
- Curation in the age of mechanical recommendations – Matt Ogle / Echo Nest – This is my pick for the Man vs. Machine talk. Matt is *the* man when it comes to understanding what is going on in the world of music listening experience.
- Crowding out the Experts – Social Taste Formation – Last.fm, Via, Rolling Stone – Is social media reducing the importance of reviewers and traditional cultural gatekeepers? Are Yelp, Twitter, Last.fm and other platforms creating a new class of tastemakers?
Music Discovery – A half dozen panels on music recommendation and discovery. Favs include:
- YouKnowYouWantIt: Recommendation Engines that Rock – Netflix, Pandora, Match.com – this panel is filled with recommendation rock stars
- The Dark Art of Digital Music Recommendations – Rovi – Michael Papish of Rovi promises to dive under the hood of music recommendation.
- No Points for Style: Genre vs. Music Networks – SceneMachine – Any talk proposal with statements like “Genre uses a 19th-century tool — a Darwinian tree — to solve a 21st-century problem. And unlike evolutionary science, it’s subjective. By the time a genre branch has been labeled (viz. “grunge”), the scene it describes is as dead as Australopithecus.” is worth checking out.
Mobile Music – Is that a million songs in your pocket or are you just glad to see me?
- Music Everywhere: Are we there yet? – Soundcloud, Songkick, Jawbone – Have we arrived at the proverbial celestial jukebox? What are the challenges?
Big Data – exploring big data sets to learn about music
- Data Mining Music – Paul Lamere – Shameless self promotion. What can we learn if we have really deep data about millions of songs?
- The Wisdom of Thieves: Meaning in P2P Behavior – Ben Fields – Don’t miss Ben’s talk about what we can learn about music (and other media) from mining P2P behavior. This talk is on my must see list.
- Big data for Everyman: Help liberate the data serf – Splunk – webifying and exploiting big data
Echo Nesty panels – proposals from folks from the nest. Of course, I recommend all of these fine talks.
- Active Listening – Tristan Jehan – Tristan takes a look at how the music experience is changing now that the listener can take much more active control of the listening experience. There’s no one who understands music analysis and understanding better than Tristan.
- Data Mining Music – Paul Lamere – This is my awesome talk about extracting info from big data sets like the Million Song Dataset. If you are a regular reader of this blog, you’ll know that I’ll be looking at things like click track detectors, passion indexes, loudness wars and son on.
- What’s a music fan worth? – Jim Lucchese – Echo Nest CEO takes a look at the economics of music, from iOS apps to musicians. Jim knows this stuff better than anyone.
- Music Apps Gone Wild – Eliot Van Buskirk – Eliot takes a tour of the most advanced, wackiest music apps that exist — or are on their way to existing.
- Curation in the age of mechanical recommendations – Matt Ogle – Matt is a phenomenal speaker and thinker in the music space. His take on the role of the curator in this world of algorithms is at the top of my SXSW panel list.
- Editor vs. Algorithm in the Music Discovery Space – SPIN, Hype Machine, Echo Nest (Jim Lucchese), 7Digital
- Defining Music Discovery through Listening – Echo Nest (Tristan Jehan), Hunted Media – This session will examine “true” music discovery through listening and how technology is the facilitator.
- Designing Future Music Experiences – Rdio, Turntable, Mary Fagot – A look at the user experience for next generation music apps.
- Music at the App Store: Lessons from Eno and Björk – Are albums as apps gimmicks or do they provide real value?
- Participatory Culture: The Discourse of Pitchfork – An analysis of ten years of music writing to extract themes.
I’ve submitted a proposal for a SXSW 2012 panel called Data Mining Music. The PanelPicker page for the talk is here: Data Mining Music. If you feel so inclined feel free to comment and/or vote for the talk. I promise to fill the talk with all sorts of fun info that you can extract from datasets like the Million Song Dataset.
Here’s the abstract:
Data mining is the process of extracting patterns and knowledge from large data sets. It has already helped revolutionized fields as diverse as advertising and medicine. In this talk we dive into mega-scale music data such as the Million Song Dataset (a recently released, freely-available collection of detailed audio features and metadata for a million contemporary popular music tracks) to help us get a better understanding of the music and the artists that perform the music.
We explore how we can use music data mining for tasks such as automatic genre detection, song similarity for music recommendation, and data visualization for music exploration and discovery. We use these techniques to try to answers questions about music such as: Which drummers use click tracks to help set the tempo? or Is music really faster and louder than it used to be? Finally, we look at techniques and challenges in processing these extremely large datasets.
- What large music datasets are available for data mining?
- What insights about music can we gain from mining acoustic music data?
- What can we learn from mining music listener behavior data?
- Who is a better drummer: Buddy Rich or Neil Peart?
- What are some of the challenges in processing these extremely large datasets?
Flickr photo CC by tristanf
I just fired up my Google Music account this afternoon and this is what I found:
All 7,861 songs are gone. I hope they come back. Apparently, I’m not the only one this is happening to.
Update – all my music has returned sometime overnight.
Peter Sobot (@psobot ) has used The Echo Nest Remix to automatically add dubstep to any song.
The Crash Bandicoot Dubset remix is pretty wild. Peter says that The Wub Machine is still work in progress. Check out how it works and add your ideas to the mix on Peter’s blog.
In a couple of weeks I’m giving a talk at SXSW called Finding Music with pictures : Data visualization for discovery. In this panel I’ll talk about how visualizations can be used to help people explore the music space and discover new, interesting music that they will like. I intend to include lots of examples both from the commercial world as well as from the research world.
I’ll be drawing material from many sources including the Tutorial that Justin and I gave at ISMIR in Japan in October 2009: Using visualizations for music discovery. Of course lots of things have happened in the year and a half since we put together that tutorial such as iPads, HTML5, plus tons more data availability. If you happen to have a favorite visualization for music discovery, post a link in the comments or send me an email: paul [at] echonest.com.
Whenever Jennie and I are in the car together, we will listen to the local Top-40 radio station (KISS 108). One top-40 artist that i can recognize reliably is Katy Perry. It seems like we can’t drive very far before we are listening to Teenage Dreams, Firework or California Gurls. That got me wondering what the average Time To Katy Perry (TTKP) was on the station and how it compared to other radio stations. So I fired up my Python interpreter, wrote some code to pull the data from the fabulous YES api and answer this very important question. With the YES API I can get the timestamped song plays for a station for the last 7 days. I gathered this data from WXKS (Kiss 108), did some calculations to come up with this data:
- Total songs played per week: 1,336
- Total unique songs: 184
- Total unique artists: 107
- Average songs per hour: 7
- Number of Katy Perry plays: 76
- Median Time between Katy Perry songs: 1hour 18 minutes
That means the average Time to Katy Perry is about 39 minutes.
Katy Perry is only the fourth most played artist on KISS 108. Here are the stats for the top 10:
|Artist||Plays|| Median time
| Average time
to next play
I took a look at some of the other top-40 stations around the country to see which has the lowest TTKP:
|Station||Songs Per Hour||TTKP|
|KIIS – LA’s #1 hit music station||8||39 mins|
|WHTZ- New York’s #1 hit music station||9||48 mins|
|WXKS- Boston’s #1 hit music station||7||39 mins|
|WSTR- Atlanta – Always #1 for Today’s Hit Music||8||38 mins|
|KAMP- 97.1 Amp Radio – Los Angeles||11||38 mins|
|KCHZ- 95.7 – The Beat of Kansas City||11||32 mins|
|WFLZ- 93.3 – Tampa Bay’s Hit Music channe||9||39 mins|
|KREV- 92.7 – The Revolution – San Francisco||11||36 mins|
So, no matter where you are, if you have a radio, you can tune into the local top-40 radio station, and you’ll need to wait, on average, only about 40 minutes until a Katy Perry song comes on. Good to know.
I was going to write a post describing all of the cool looking music-oriented panels that have been proposed for SXSW 2011, but debcha at zed equals zee beat me to it. Be sure to read Deb’s SXSWi 2011 panel proposals in music and tech post. Some of the panels I’m looking to the most are:
Digital Music Smackdown: The Best Digital Music Service – In what is expected to be a heated and fiercely competitive discussion, C and VP-level executives from four digital music companies (MOG, Spotify, Pandora and Rhapsody) battle it out over the title of “Best Digital Music Service. This could be fun if it is really a smackdown, but I suspect that the execs will be very polite and complimentary of each other’s services leading to a boring panel. I hope I’m wrong. Also, where’s Last.fm? – they should be on the panel too.
We Built this App on RocknRoll: Style Matters – For an inherently auditory medium, music is ingrained with style. From 12″ artwork and niche mp3 blogs to the latest design on your sweatshirt or skate deck, music has always been analogous with visual culture. So what happens when you overlay this complex fabric of cultural values and personal identities on what is already a thorny process: building and launching a music app. – Hannah of Last.fm and Anthony of Hype Machine talk about the design of music apps. These two know their stuff. Should be really interesting.
Music & Metadata: Do Songs Remain the Same? Metadata may be an afterthought when it comes to most people’s digital music collections, but when it comes to finding, buying, selling, rating, sharing, or describing music, little matters more. Metadata defines how we interact and talk about music—from discreet bits like titles, styles, artists, genres to its broader context and history. Metadata builds communities and industries, from the local fan base to the online social network. Its value is immense. But who owns it? This panel is on my Must See list.
Expressing yourself musically with Mobile Technology – This is a panel with Ge Wang, founder/CTO of Smule talking about creating music on mobile devices. Ge is an awesome speaker and gives great demo. Don’t miss this one.
Music APIs – A Choreographed Dance with Devices? – This panel discussion focuses on real-world examples beyond the fundamentals or technical aspects of an API. Attend this panel and review success stories from the pros that demonstrate how an API brings content, software, and hardware together. Looks like a good Music APIs 101 for biz types.
I would be remiss if I didn’t pimp my own panels. Be sure to consider (and maybe even comment on / vote for ) these panels:
Love, Music & APIs. – Consider this to be the Music Hack Day panel. Dave Haynes (SoundCloud) and I will talk about the impact that Music APIs are having on the world of music and how programmers will soon be the new music gamekeeper.
Finding Music With Pictures: Data Visualization for Discovery: In this panel I’ll look at how visualizations can be used to help people explore the music space and discover new, interesting music that they will like. We will look at a wide range of visualizations, from hand drawn artist maps, to highly interactive, immersive 3D environments.
The folks at SXSW are looking for input on these panels to help decide what makes it onto the schedule, so if any of these strike your fancy, head on over to the panel descriptions and add your comments.
Next week I’ll be giving a talk about remixing music with Echo Nest remix at the Boston Python Meetup Group. If you are in the Boston / Cambridge area next week, be sure to come on by and say ‘hi’. Info and RSVP for the talk are here: The Boston Python Meetup Group on Meetup.com
Here’s the abstract for the talk:
Paul Lamere will tell us about Echo Nest remix. Remix is an open source Python library for remixing music. With remix you can use Python to rearrange a track, combine it with others, beat/pitch shift it etc. – essentially it lets you treat a song like silly putty.
The Swinger is an interesting example of what it can do that made the rounds of the blogosphere: it morphs songs to give them a swing rhythm.
For more details about the type of music remixing you can do with remix, feel free to read: http://musicmachinery…
[tweetmeme source= ‘plamere’ only_single=false]
TL;DR; I built a game called Name Dropper that tests your knowledge of music artists.
One bit of data that we provide via our web APIs is Artist Familiarity. This is a number between 0 and 1 that indicates how likely it is that someone has heard of that artists. There’s no absolute right answer of course – who can really tell if Lady Gaga is more well known than Barbara Streisand or whether Elvis is more well known than Madonna. But we can certainly say that The Beatles are more well known, in general, than Justin Bieber.
To make sure our familiarity scores are good, we have a Q/A process where a person knowledgeable in music ranks our familiarity score by scanning through a list of artists ordered in descending familiarity until they start finding artists that they don’t recognize. The further they get into the list, the better the list is. We can use this scoring technique to rank multiple different familiarity algorithms quickly and accurately.
One thing I noticed, is that not only could we tell how good our familiarity score was with this technique, this also gives a good indication of how well the tester knows music. The further a tester gets into a list before they can’t recognize artists, the more they tend to know about music. This insight led me to create a new game: The Name Dropper.
The Name Dropper is a simple game. You are presented with a list of dozen artist names. One name is a fake, the rest are real.
If you find the fake, you go onto the next round, but if you get fooled, the game is over. At first, it is pretty easy to spot the fakes, but each round gets a little harder, and sooner or later you’ll reach the point where you are not sure, and you’ll have to guess. I think a person’s score is fairly representative of how broad their knowledge of music artists are.
The biggest technical challenge in building the application was coming up with a credible fake artist name generator. I could have used Brian’s list of fake names – but it was more fun trying to build one myself. I think it works pretty well. I really can’t share how it works since that could give folks a hint as to what a fake name might look like and skew scores (I’m sure it helps boost my own scores by a few points). The really nifty thing about this game is it is a game-with-a-purpose. With this game I can collect all sorts of data about artist familiarity and use the data to help improve our algorithms.
So go ahead, give the Name Dropper a try and see if you can push me out of the top spot on the leaderboard: