Archive for category The Echo Nest
My Music Hack Day Berlin hack was “Where’s the Drama?” – a web app that automatically identifies the most dramatic moment in any song and plays it for you. I’ve been having lots of fun playing with it … and even though (or perhaps because) I know how it works, I’m often surprised at how well it does at finding the most dramatic moments. Here are some examples:
- When will the Bass Drop - Lonely Island
- Stairway to Heaven – Led Zeppelin
- Doomsday – Nero
- November Rain - Guns N Roses
How does it work? The app grabs the detailed audio analysis for the song from The Echo Nest. This includes a detailed loudness map of the song. This is the data I use to find the drama. To do so, I look for the part of the song with the largest rise in volume over the course of a 30 second window (longer songs can have a bit of a longer dramatic window). I give extra weight to crescendos that culminate in louder peaks (so if there are two crescendos that are 20dB in range but one ends at 5dB louder, it will win). Once I identify the most dynamic part of a song, I pad it a bit (so we get to hear a bit of the drop after the build).
Playing the music - I wanted to use Spotify to play the music, which was a bit problematic since there currently isn’t a way to play full streams with the Spotify Web API, so I did a couple of hacky hacks that got me pretty far. First of all, I discovered that you can add a time offset to a Spotify URI like so:
When this URI is opened in Spotify (even when opened via a browser), Spotify will start to play the song a the 1:05 time offset.
I still needed to be able to stop playing the track – and there’s no way to do that directly – so instead, I just open the URI:
which happens to be the URI for John Cage’s 4’33. In other words, to stop playing one track, I just start playing another (that happens to be silent). The awesome side effect of this is that I’ll be slowly turning anyone who uses “Where’s the Drama?” into experimental music listeners as the Spotify recommendation system responds to all of those John Cage ‘plays’. This should win some sort of ‘hackiest hack of the year’ award.
It was a fun hack to make, and great fun to demo. And now that I have the app, I am no longer wasting time listening to song intros and outros, I can just get to the bit of the song that matters the most.
For my summer vacation early-morning coding for fun project I revamped my old Acrostic Playlist Maker to work with Spotify. The app, called Acrostify, will generate acrostic playlists with the first letter of each song in the playlist spelling out a secret message. With the app, you can create acrostic playlists and save them to Spotify.
Give it a try at Acrostify.
I am at Outside Hacks this weekend – A hackathon associated with the Outside Lands music festival. For this hack I thought it would be fun to try out the brand new Your Music Library endpoints in the Spotify Web API. These endpoints let you inspect and manipulate the tracks that a user has saved to their music. Since the hackathon is all about building apps for a music festival, it seems natural to create a web app that gives you festival artist recommendations based upon your Spotify saved tracks. The result is the Outside Lands Recommender:
The Recommender works by pulling in all the saved tracks from your Spotify ‘Your Music’ collection, aggregating the artists and then using the Echo Nest Artist Similar API to find festival artists that match or are similar to those artists. The Spotify API is then used to retrieve artist images and audio samples for the recommendations where they are presented in all of their bootstrap glory.
This was a pretty straight forward app, which was good since I only had about half the normal hacking time for a weekend hackathon. I spent the other half of the time building a festival dataset for hackers to use (as well as answering lots of questions about both the Spotify and Echo Nest APIs).
It has been a very fun hackathon. It is extremely well organized, the Weebly location is fantastic, and the quality of hackers is very high. I’ve already seen some fantastic looking hacks and we are still a few hours from demo time. Plus, this happened.
Last month I released the Set Listener – a web app that lets you create a Spotify playlist for an artist’s most recent show. A frequent request by users has been to allow the creation of a Spotify playlist for any show, not just the most recent one. However, I didn’t want to have to implement an elaborate event search and browse feature (this was supposed to be a quick hack) especially since SetList.fm already implements this browsing. So instead I did a bit of hack to implement this feature – in addition to accepting an artist name, the Set Listener will accept the url of any setlist.fm show. If you enter a setlist.fm show URL, the Set Listener will grab the songs for that show and generate the playlist. Here’s the flow:
Go to setlist.fm and browse until you find your show of interest. Grab the URL
Paste the URL into the Set Listener and you should be good to go:
I am wearing my International Executive Music Hacker hat today. I’m writing this blog post at 5AM somewhere over the Atlantic Ocean, on my way to the Barcelona Music Hack Day, where I’ll be representing both The Echo Nest and Spotify. I’m pretty excited about the hack event – first, because it’s in freaking Barcelona, and second, because I get to talk about what’s been going on with the Spotify and Echo Nest APIs.
It has been just about 100 days since The Echo Nest and Spotify have joined forces. In that time we’ve been working hard to build the best music platform for listeners and for developers. This week we are releasing some of the very first fruits of our labors.
First up, we are releasing a new Spotify Web API.
This is a complete revamp of the Spotify Metadata API (the old version has now been deprecated). The Spotify Web API gives you access to all sorts of information about the Spotify catalog including details about artists, albums and tracks. Want to know the top tracks for an artist? There’s an API for that. Looking for high quality album art, artist images and 30 second audio previews? There are APIs for that too. Best of all, the new API includes perhaps the most requested Spotify API feature of all time With the Spotify Web API you can now create and modify playlists on behalf of authenticated users. Yes – you can now create a Spotify web app that creates playlists. (I personally requested this feature way back in 2008, here’s my begging plea for the feature in 2009).
I’ve been using the beta version of this new API for a couple months now and I must say I am quite impressed. The API is fast, super easy to use, and provides all sorts of great data for building apps. In the past weeks I’ve had fun converting a number of my favorite apps to use the Spotify API. First there’s the Road Trip Mix Tape that lets you create a Spotify playlist of music by artists that are from the very towns you are driving through. Then there’s Music Popcorn, a visual interface for exploring genres. For the less visual, there’s the Genre Browser that gives you lots of details about the different music genres including playlists that help give you a gentle introduction to any of the thousands of Echo Nest genres. Next there’s Boil the Frog, an app that creates seamless playlists between any two artists. Finally there’s the 3D Music Maze, an app that lets you explore for music by wandering through a 3 dimensional music world.
Next up, a freshly minted Echo Nest + Spotify Sandbox — a new Spotify ID space.
These apps are possible because of the second thing we are releasing this week – a spiffy, shiny new Spotify Rosetta Stone catalog that ensures that the Echo Nest API has the freshest, and most up-to-date view of the Spotify universe of music. For those who might be new to The Echo Nest, Project Rosetta Stone is something we’ve been working on here at the Nest for many years. The goal of Project Rosetta Stone is to solve one of the most common problems that nearly every music app developer faces. The problem is that every music service has its own set of IDs – a music subscription service like Spotify has its own artist, album and track IDs. A lyric service has its own (and very different) IDs for those same artists, albums and tracks and a concert ticketing API has yet a third set of IDs. This is quite problematic for app developers that want to build an app that combines information from multiple services. Without a common ID system, the app developer has to resort to metadata searching and matching – which is slow and quite error prone – this results in a poor app.
Project Rosetta Stone solves this problem by providing ID mappings between as many music services as we can. With this mapping you can easily translate IDs from one ID space to another. With Rosetta Stone, if you have the Spotify track ID you can get Lyricfind and/or Musixmatch IDs making it easy to use those respective APIs to retrieve lyrics for that song. You can easily map the Spotify artist ID to a Songkick or Eventful ID to get ticket and touring information from those APIs. And of course you can use the Spotify track ID to get detailed Echo Nest information about the song such as its tempo, energy, danceability, along with detailed Echo Nest artist data such as latest artist news, blog posts and similar artists.
We have had Spotify IDs in Rosetta Stone for many years, but this particular mapping has in the past been problematic for us. Spotify has a huge catalog and keeping the mapping fresh and up to date between Spotify and The Echo Nest has always been a big challenge. There’s a huge back catalog with millions of tracks to deal with plus thousands of new tracks are being added every week. The result was that there was always a bit of a lag between when updates to the Spotify catalog were reflected in the Rosetta Stone mapping. This meant that if you built a Rosetta Stone-based app you could find that The Echo Nest wouldn’t always know about a Spotify track, especially if a track was very new. The result would be a less-than-perfect app.
This week we are releasing a new Spotify ID space. Our engineers have been working hard over the last 100 days to set up all sorts of infrastructure and plumbing to ensure that we have the most up-to-date view of the Spotify catalog. No more lag between when a new track appears in Spotify and when you can get Echo Nest data. Plus, all of our APIs that take IDs as inputs will now also take Spotify IDs as input as well. If you have a Spotify artist ID you can use it with any Echo Nest artist API method. Likewise, if you have a Spotify track ID you can use it with any Echo Nest song or track API method that takes a track ID as input. This makes it **really** easy for developers to use The Echo Nest and Spotify Apps together. For example, here’s an API call that returns detailed audio properties for a Spotify track given its ID.
I’ve been having much fun using The Echo Nest API with the brand new Spotify API. I’ve already written some code that you can use. First, I wrote a Python library for Spotify called Spotipy. It’s makes it easy to write Python programs that use the new Spotify Web API, and it works well with my Echo Nest Python library called Pyen. Here’s an example of using the two libraries together:
So yes, I’m pretty jazzed about this trip to Barcelona. I get to create a music hack, I get to spend a few days with some of the best music hackers in the world (The Barcelona Music Hack Day, as part of the Sonar Festival tends to attract the top music hackers). I get to spend a few days on the Mediterranean in one of the most beautiful cities in the world. Best of all, I get to talk about the new Spotify and Echo Nest developer platform and help music hackers build cool stuff on top of the newly combined platform.
I’ve put together a page that talks in detail about the new Spotify / Echo Nest platform. It has links to all of the API docs, libraries, examples, github repos, demos and details on how you can use The Echo Nest / Spotify Platform. Check it out here:
Keep an eye on this space for I’ll be updating it as we continue to integrate our developer APIs. There’s lots more coming, so stay tuned!
Ten years ago, May 3, 2004 I typed my very first blog post. At the time I was a researcher at Sun Microsystem’s Lab in the speech group working on speech synthesis and speech recognition systems, and so that was what I blogged about. Over the next year, my blog, first called “Duke Speaks!” and then called “Duke Listens!”, (Duke is the name of the Java mascot) slowly morphed into a blog that was focused on music technology, and in particular, music information retrieval, music recommendation, music playlisting and discovery. Five years later, when I left Sun to join The Echo Nest, I relaunched the blog as Music Machinery where I’ve been writing ever since.
I’ve written lots of blog posts. Some of my favorite from the Duke Listens! days are:
- My first MIR-related post (June 2004)
- My first hardcore MIR post (January 2005)
- A very wrong prediction about Apple (January 2005)
- I discover Radio Paradise (April 2005)
- First Google Music rumor (June 2005)
- First Amazon Music rumor (August 2005)
- First Pandora Post (September 2005)
- First mention of The Echo nest (October 2005)
- Why there’s no Google Music search (December 2005)
- First mention of Spotify (January 2007)
- My review of Spotify (November 2007)
- The Echo Nest goes live (March 2008)
- The Echo Nest launches their API (September 2008)
- My first look at iTunes genius recommendations (September 2008)
- My last post (February 2009)
Some of my more popular posts from the Music Machinery days are:
- The Swinger
- Inside the precision hack
- Exploring regional listening preferences
- moot wins, Time Inc. loses
- The Infinite Jukebox
- In search of the click track
- The Loudness War Analyzed
- Hacking spotify
- Favorite Artists vs Distinctive Artists by State
- How good is Google’s Instant Mix?
- How to process a million songs in 20 minutes
- Joco vs. Glee
- Precision Hacking
I’m really looking forward to the next decade of blogging. Now that I work for Spotify, there’s a seemingly unlimited supply of music data that will be prime blogging material. It will be great fun.
Thanks to you my reader, for stopping by, and for all of your great comments, and feedback over the years.
At the time when I was coming of age musically, when we listened to music on LPs, the listening experience was very different than it is today. For one, if you didn’t like the currently playing song you had to get out of your chair, walk over to the turntable, carefully pick up the tone arm and advance the needle to the next track. That was a lot of work to avoid three minutes of bad music. You really had to really dislike a song to make skipping it worth the effort. Today, with our fancy iPhones and our digital streaming music subscription services, skipping a song couldn’t be easier. Just tap a button and you are on to the next song. The skip button is now a big part of the overall listening experience. Don’t like a song? Skip it. Never heard a song? Skip it. Just heard a song? Skip it. The Skip even plays a role in how we we pay for music. For most music subscription services if you want the freedom to skip a song whenever you want, you’ll need to be a premium subscriber, otherwise you’ll be limited to a half-dozen or so skips per hour.
I am interested in how people are using the skip button when listening to music so I spent a bit of time taking a closer look at skip data. This and the next blog post or two will be all about the skipping behavior of music listeners. We’ll take a look at how often people skip, whether different listener demographics have different skipping behavior, what artists and genres are most and least likely to trigger skips and more!
This is my first deep dive into Spotify data. The Spotify team has built up a fantastic big data infrastructure making it easy to extract insights from the billions and billions of music plays. For this study I’ve processed several billions of plays from many million unique listeners from all around the world.
What is a skip?
For this study, I define a skip as any time the listener abandons a song before the song finishes. It could be because the listener explicitly presses the skip button, or it could be that they searched for and started another song before the current song finished, or they clicked on a different song in the playlist. For whatever reason, if the listener doesn’t make it to the end of the song, I call it a skip.
How often do people skip?
The first and most basic question to answer is: How often do people skip?. Given that skipping is so easy how big of a part does skipping play in our listening. The answer: A lot!
Here are the numbers. First, lets look at how often a song is skipped within the first five seconds of play. I call these quick skips. The likelihood that a song will be skipped within the first five seconds is an astounding 24.14%. Nearly one quarter of all song plays are abandoned in the first 5 seconds. The likelihood that a song will be skipped within the first thirty seconds rises to 35.05%. The chance that a song is skipped before it ends is a whopping 48.6%. Yes, the odds are only slightly better than 50/50 that a song will be played all the way to the end.
|Skipped in||Likelihood of skip|
|First 5 seconds||24.14 %|
|First 10 seconds||28.97 %|
|First 30 seconds||35.05 %|
|Before song finishes||48.6 %|
The following plot shows the average skipping behavior for millions of listeners and billions of plays. The plot shows the rather steep drop off in listeners in the early part of a song when most listeners are deciding whether or not to skip the song. Then there’s a slow but steady decline in listeners until we reach the end of the song where only about 50% of the listeners remain.
The next plot shows the average skipping behavior within in the first 60 seconds of a song. It shows that most of the song skips happen within the first 20 seconds or so of the song, and after that there’s a relatively small but steady skipping rate.
We can also calculate an overall skip rate per listener - that is, the average number of times a listener skips a song per hour.
Average listener/skips per hour: 14.65
On average a listener is skipping a song once every four minutes. That’s a whole lot of skipping.
Who is doing all that skipping?
Do different types of listeners skip music at different rates? Lets take a look.
Skipping rate of male listeners: 44.75%
Skipping rate of female listeners: 45.23%
There seems to be little difference as to how often men and women skip.
Desktop skipping rate: 40.1%
Mobile skipping rate: 51.1%
When we are at our desktops, we tend to settle into longer listening sessions and skip less, while when we are mobile we spend much more time interacting with our music.
This plot shows the skipping rate as a function of the age of the listener. It shows that young teenagers have the highest skipping rate – well above 50%, but as the listener gets older their skipping rate drops rather dramatically, to reach the skipping nadir of about 35%. Interestingly, the skipping rate rises again for people in their late 40s and early 50s. I have a couple of theories about why this might be. The first theory is that the skipping rate is a indication of how much free time a person has time. Teenagers skip more because they have more time to devote to editing their music stream, whereas thirty-somethings, with their little kids and demanding jobs, have no time to pay attention to their music players. The second theory, suggested by Spotify über-analyst Chris Tynan, is that the late-forties skipping resurgence is caused by teenagers that use their parent’s account.
When do people skip the most?
The following plot shows the skipping behavior over a 24 hour period. To create the plot, I analyzed the listening behavior for UK residents (which are conveniently confined to a single timezone) over several weeks.
The plot shows that the skipping rate is lowest when people are paying less attention to music – like when they are asleep, or at work. Skipping behavior peaks in the morning hour as people start they day and start to head into work and again at the end of the day when they are at home or out socializing with their friends. The plot shows the time of day when people tend to have the most attention to devote to hand-curating their music stream. When people are sleeping or working, their skip rate goes down.
In the next plot, below, the skipping rate is overlaid with normalized song plays. It is interesting to see that the highest skipping rates do not coincide with the peak music playing times of the day, but instead is aligned with the times of day when rate of change in plays is the most.
Skipping behavior by Day of the Week
The following plot shows the average skipping rate per day of the week. The skipping rate is higher on weekends, showing, once again, that when people have more spare time, they are more apt to curate their listening sessions by skipping tracks.
The Skip really has changed how we listened to music. It plays a significant role in how we interacts with our music stream. When we are more engaged with our music – we skip more, and when music is in the background such as when we are working or relaxing, we skip less. When we have more free time, such as when we are young, or on the weekends, or home after a day of work, we skip more. That’s when we have more time to pay attention to our music. The big surprise for me is how often we skip. On average, we skip nearly every other song that we play.
Skipping has become an important part of how we listen to music. It is no surprise then, that ‘unlimited skipping’ is a feature used to entice people to upgrade to a premium paid account. And it may be one of the reasons why people would switch from a service that doesn’t offer unlimited skips even on their premium service to one that does.
Coming soon: Look for my next post that will look at which genres, songs and artists get skipped the most and the least.