Archive for February, 2009
The Echo Nest Remix SDK
Posted by Paul in fun, Music, The Echo Nest on February 28, 2009
One of the joys of working at the Echo Nest is the communal music playlist. Anyone can add, rearrange or delete music from the queue. Of course, if you need to bail out (like when that Cindi Lauper track is sending you over the edge) you can always put on your headphones and tune out the mix. The other day, George Harrison’s “Here Comes the Sun” started playing, but this was a new version – with a funky drum beat, that I had never heard before – perhaps this was a lost track from the Beatle’s Love? Nope, turns out it was just Ben, one of the Echo Nest developers, playing around with The Echo Nest Remix SDK.
The Echo Nest Remix SDK is an open source Python library that lets you manipulate music and video. It sits on top of the Echo Nest Analyze API, hides all of the messy details of sending audio back to the Echo Nest, and parsing the XML response, while still giving you access to the full power of the API.
remix – is one of The Echo Nest’s secret weapons – it gives you the ability to analyze and manipulate music – and not just audio manipulations such as filtering or equalizing, but the ability to remix based on the hierarchical structure of a song. remix sits on top of a very deep analysis of the music that teases out all sorts of information about a track. There’s high level information such as the key, tempo time signature, mode (major or minor) and overall loudness. There’s also information about the song structure. A song is broken down into sections (think verse, chorus, bridge, solo), bars, beats, tatums (the smallest perceptual metrical unit of the song) and segments (short, uniform sound entities). remix gives you access to all of this information.
I must admit that I’ve been a bit reluctant to use remix – mainly because after 9 years at Sun Microsystems I’m a hard core Java programmer (the main reason I went to Sun in the first place was because I liked Java so much). Every time I start to use Python I get frustrated because it takes me 10 times longer than it would in Java. I have to look everything up. How do I concatenate strings? How do I find the length of a list? How do I walk a directory tree? I can code so much faster in Java. But … if there was ever a reason for me to learn Python it is this remix SDK. It is just so much fun – and it lets you do some of the most incredible things. For example, if you want to add a cowbell to every beat in a song, you can use remix to get the list of all of the beats (and associated confidences) in a song, and simply overlap a cowbell strike at each of the time offsets.
So here’s my first bit of Python code using remix. I grabbed one of the code samples that’s included in the distribution, had the aforementioned Ben spend two minutes walking me through the subtleties of Audio Quantum and I was good to go. My first bit of code just takes a song and swaps beat two and beat three of all measures that have at least 3 beats.
def swap_beat_2_and_3(inputFile, outputFile):
audiofile = audio.LocalAudioFile(inputFile)
bars = audiofile.analysis.bars
collect = audio.AudioQuantumList()
for bar in bars:
beats = bar.children()
if (len(beats) >= 3):
(beats[1], beats[2]) = (beats[2], beats[1])
for beat in beats:
collect.append(beat);
out = audio.getpieces(audiofile, collect)
out.encode(outputFile)
The code analyzes the input, iterates through the bars and if a bar has more than three beats, swaps them. (I must admit, even as a hard core Java programmer, the ability to swap things with (a,b) = (b,a) is pretty awesome) and then encodes and writes out a new audiofile. The resulting audio is surprisingly musical. Here’s the result as applied to Maynard Ferguson’s “Birdland”:
This is just great programming fun. I think I’ll be spending my spare coding time learning more Python so I can explore all of the things one can do with remix.
Hacking spotify
Spotify is the new “old napster” – everyone who uses it seems to love it. As this Google trends plot shows it is starting to become very popular.
But there is a downside to becoming popular – when you are popular you start to become a target of hackers. This is happening to Spotify now – Spotify is another platform waiting to be explored and exploited. Some notable hacks:
- Lastify – this is a rather benign hack – it adds a couple of buttons to the bottom of your spotify client that let you apply Last.fm ‘love’ and ‘ban’ to the currently playing track.
- Despotify – the open source Spotify client – this is a rather extensive hack. #hack.se has reverse engineeered the Spotify protocols and have built an open source Spotify client (with curses text-mode goodness). The client includes code that decrypts the encrypted music served by Spotify, potentially allowing anyone to not just listen to music, but to download and save it as well. Here’s a video of Despotify in action:
Already, Spotify seems to have responded to this hack, according to the Despotify page: “Despotify has been blocked for users using ‘free’ or ‘daypass’ accounts. You can still use despotify using ‘Premium’ accounts.”. That seems fair – if you pay for Spotify, you can use whatever client you want.
- Geographic hacks – Spotify is only released in certain countries. If you don’t live in the UK, Spain, France, Sweden, Norway or Finland you are out of luck – but not really. According to this article in Wired, some users are using a UK-based proxy to allow access to Spotify from places like the USA.
As Spotify gains in popularity, the Spotify engineers are going to be playing a bit of wack-a-mole to keep the hackers at bay in order to keep the Spotify platform stable and performant. So far, they seem to be doing a very good job.
setlist.fm – the setlist wiki
setlist.fm is a wiki-like service where people can record and share the setlists for concerts they’ve attended. Interested in learning what Yes might play should you see them, you can look at the setlist for their recent concert in Georgia:
Firebird Suite
Siberian Khatru
I’ve Seen All Good People
Tempus Fugit
Onward
Astral Traveler
Close To The Edge
J’s Theme
Intersection Blues
And You And I
Long Distance Runaround
The Fish (Schindleria Praematurus)
Aliens (Are Only Us From The Future)
Machine Messiah
Starship Trooper
Owner Of A Lonely Heart
Roundabout
setlist.fm doesn’t just show you the setlist, it also creates links to Youtube videos for each of the tracks, finds the lyrics from the LyricWiki. setlist calculates nifty statistics about which songs a band has played most in their concerts. Setlist.fm is a neat idea – and the site design and implementation is really slick. It’s a pretty cool site.
The Led Zeppelin Graph
Posted by Paul in fun, Music, The Echo Nest on February 26, 2009
I’ve been pretty busy figuring out the lay of the land at the new job, so I haven’t had too much time for recreational programming. However, last night, while my lovely wife was watching Dr. House demonstrate his excellent interpersonal skills, I got a chance to write a little bit of code to generate an artist graph using the Echo Nest Developer API.
The idea is to generate a graph that shows the artist similarity space in a fashion that can encourage exploration of the artist space. To do this, I simply use the Echo Nest get_similar call to walk the artist graph. Instead of getting bogged down in some graphics library to create the visualization, I just output ‘.DOT’ commands and render the whole thing using graphviz. Graphviz does all the hard work figuring out how to layout the graph. Here’s a tiny example of some graphviz output:
One of the problems with making these sort of graphs is that they can get extremely complicated, very quickly. Even after just a few steps away from the seed artist in the crawl of the artist graph there can be 100s of artists and 1000s of connections. Without some care, the graph quickly turns into an unreadable tangle. However, since we want to use these graphs for exploration of the artist space we can make a simplification that eliminates much of the complexity. For exploration, people tend to start from a known artist, and then proceed to lesser known artists. If we make our graph work in the same way, we will eliminate a large number of extraneous connections. Instead of connecting all artists that we encounter in our crawl of the artist graph, we only connect new artists to more popular artists that are already in the graph. This gives us an easy to manage directed, acyclic graph that flows from very familiar artists to unknown artists.
The pseudocode to do this is very simple:
add a seed artist to the work queue
while the work queue is not empty
curArtist <= the next artist from the queue
for each artist similar to curArtist
if similar artist less familiar than curArtist
plot link to similar artist
add similar artist to workqueue
The real java code is not much more complicated:
while (workQueue.size() > 0) {
Artist artist = workQueue.remove(0);
List<Scored<Artist>> simArtists = echoNest.getSimilarArtists(artist, 0, 6);
float familiarity = echoNest.getFamiliarity(artist);
for (Scored<Artist> scoredArtist : simArtists) {
Artist similarArtist = scoredArtist.getItem();
float simFamiliarity = echoNest.getFamiliarity(similarArtist);
if (simFamiliarity < familiarity) {
out.printf("\"%s\" -> \"%s\";\n", artist.getId(), similarArtist.getId());
if (!plottedSet.contains(similarArtist)) {
workQueue.add(similarArtist);
plottedSet.add(similarArtist);
out.printf("\"%s\" [label=\"%s\"]\n", similarArtist.getId(), similarArtist.getName());
}
}
}
}
This yields some fun graphs. Here’s a detail from a graph created using Led Zeppelin as the see artist:
And the full graph in all its glory is here:
Full plot (click to see it full size)
I can think of all sorts of things to add to this artist graph. We could size the nodes based upon the familiarity of the artist. We could color the artists based upon how ‘hot‘ the artist is. We could replace the graphviz with a real graphing library like prefuse and make the whole graph interactive – so you could actively explore the artist space, click on a node, read reviews about the artist, listen to their music, watch their videos.
Astute readers may have noticed that I’m making calls using an EchoNest library. That’s one of the things I’ve been working on in the last week – building a Java client library for the EchoNest developer API. I’ll be releasing this soon, once I figure out the best way to release an open source client library here at The Echo Nest. I should hopefully get something released by the end of this week. If you are interested in a sneak preview of the Java client library, let me know.
Last.fm and the iPhone
Here’s a nifty iPhone commercial that highlights Last.fm that has been running in the UK. Cool stuff, nicely done Toby!
One Blog, Two Blog, Old Blog, New Blog
Here are a couple of blogs to add to your blog roll. First, Stephen Green (aka SearchGuy) has started posting to his blog again. Steve writes indepth articles about the innards of a search engine – and why that inverted text file that you created for your CS 301 homework is not going to put Google out of business anytime soon. It’s a good blog: SearchGuy.
Second, Jeremy seems to now be blogging – this makes me quite sad, because Jeremy has regularly emailed me blog fodder – so now that he has his own blog, I suspect that source will dry up. But it is all for the greater good. Jeremy is writing interesting articles about search from a higher vantage point than Steve. Jeremey says: “My idea was to have a place where interested researchers and search observers can gather, survey, and discuss information retrieval from a useful vantage point: somewhere tall where you can get a good overview of what is happening.” Jeremy is blogging at Information Retrieval Gupf.
The Billboard API
Posted by Paul in Music, web services on February 24, 2009
Billboard, the venerable maintainer of the Billboard Hot 100 and a bevy of other music charts, is now making this data available via an API. The API “puts the entire rich history of the Billboard charts at your fingertips to sample and mix into your web pages and applications.”. The API is in public beta – but already it is supplying some really good information.
The first service that they’ve rolled out is the ‘Chart’ service, which lets you search and retrieve Billboard chart information.
For example, to find all appearances of The Beatles on any of the Billboard charts during the first week of June in 1964, you could make the call:
With results:
<?xml version='1.0' encoding='UTF-8'?>
<searchResults firstPosition='1' totalReturned='6' totalRecords='6'>
<chartItem id='8807769' rank='2' exrank='0'>
<chart id='3070264'>
<name>The Billboard Hot 100</name>
<issueDate>1964-06-06</issueDate>
<specId>379</specId>
<specType>Singles</specType>
</chart>
<artist>The Beatles</artist>
<writer />
<song>Love Me Do</song>
<producer />
<catalogNo>9008</catalogNo>
<promotion />
<distribution>Tollie</distribution>
<peak>1</peak>
<weeksOn>14</weeksOn>
</chartItem>
<chartItem id='8715479' rank='4' exrank='0'>
<chart id='3068613'>
<name>The Billboard 200</name>
<issueDate>1964-06-06</issueDate>
<specId>305</specId>
<specType>Albums</specType>
</chart>
<artist>The Beatles</artist>
<writer />
<song>The Beatles' Second Album</song>
<producer />
<catalogNo>2080</catalogNo>
<promotion />
<distribution>Capitol</distribution>
<peak>1</peak>
<weeksOn>55</weeksOn>
</chartItem>
<chartItem id='8715481' rank='6' exrank='0'>
<chart id='3068613'>
<name>The Billboard 200</name>
<issueDate>1964-06-06</issueDate>
<specId>305</specId>
<specType>Albums</specType>
</chart>
<artist>The Beatles</artist>
<writer />
<song>Meet The Beatles!</song>
<producer />
<catalogNo>2047</catalogNo>
<promotion />
<distribution>Capitol</distribution>
<peak>1</peak>
<weeksOn>71</weeksOn>
</chartItem>
<chartItem id='8807803' rank='36' exrank='0'>
<chart id='3070264'>
<name>The Billboard Hot 100</name>
<issueDate>1964-06-06</issueDate>
<specId>379</specId>
<specType>Singles</specType>
</chart>
<artist>The Beatles</artist>
<writer />
<song>Do You Want To Know A Secret</song>
<producer />
<catalogNo>587</catalogNo>
<promotion />
<distribution>Vee-Jay</distribution>
<peak>2</peak>
<weeksOn>11</weeksOn>
</chartItem>
<chartItem id='8715486' rank='11' exrank='0'>
<chart id='3068613'>
<name>The Billboard 200</name>
<issueDate>1964-06-06</issueDate>
<specId>305</specId>
<specType>Albums</specType>
</chart>
<artist>The Beatles</artist>
<writer />
<song>Introducing...The Beatles</song>
<producer />
<catalogNo>1062</catalogNo>
<promotion />
<distribution>Vee-Jay</distribution>
<peak>2</peak>
<weeksOn>49</weeksOn>
</chartItem>
<chartItem id='8807777' rank='10' exrank='0'>
<chart id='3070264'>
<name>The Billboard Hot 100</name>
<issueDate>1964-06-06</issueDate>
<specId>379</specId>
<specType>Singles</specType>
</chart>
<artist>The Beatles</artist>
<writer />
<song>P.S. I Love You</song>
<producer />
<catalogNo>9008</catalogNo>
<promotion />
<distribution>Tollie</distribution>
<peak>10</peak>
<weeksOn>8</weeksOn>
</chartItem>
</searchResults>
You can restrict searches to various charts (Hot Country, Pop 100, Top Latin, etc.) , and you can search by artist and/or song name over a range of dates. (Unfortunately, but not too surprisingly, the data for the current month is not available in the searches).
The terms-of-service seem pretty reasonable- you are allowed to make 1,500 API calls per day at up to 2 queries per second. Commercial use seems to be allowed (But I’m not a lawyer, so you should check for yourself). However, according to the terms, you are not allowed to store any of the Billboard data. The services are well documented, support JSON as well as XML output and query times are fast.
I can think of all sorts of uses for this data – to help create playlists for the 25 year high school reunion, tracking artist popularity over time, answering bar room music questions like “What was the highest charting instrumental-only single?” or “Did Ringo ever have a hit?”. It is perfect data for the Music Alchemists that are trying to build an automatic hit predictor.
The Billboard chart API is an excellent addition to the world of music web services. It goes straight into my Top Ten Music APIs chart – with a bullet.
Collaborative Playlists
Posted by Paul in Uncategorized on February 23, 2009
Collaborative playlists seem to be all the rage. Fred Wilson writes how he collaborated with his twitter followers to create a primal screen playlist. Jason talks about combining twitter search, hashtags and Lucas’s playTwitter. Twisten and Blip let you twitter songs and play tweets of others. But I’ve been creating and enjoying collaborative playlists for over a year with Spotify. Collaborative playlists are fun and a great way to share music and Spotify’s approach makes it dead easy. You just create a playlist (easy as iTunes), mark it as ‘collaborative’ and share the URL of the playlist with your friends (or the world). Anyone with the URL to the shared playlist can add to, delete or reorder songs on the playlist. Of course, the down side is that someone can really mess up your playlists. I have a few suggestions that could make Spotify collaborative playlists even better:
- Add version control to playlists – so when some vandal adds “Never gonna give you up” 50 times to your playlist you can recover
- Allow playlist editors to add tags or notations to their additions so you can see why a particular song was added to the playlist
- Allow for lengthy text description of playlists and tracks in the playlist (like the XSPF supports)
- Let me share a read-only version of a collaborative playlist
- Create a directory of playlists so that Spotify users can easily find public and collaborative playlists by name, tag or description.
One day perhaps the whole world will be able to enjoy Spotify’s collaborative playlists.
A recommender comic …
Posted by Paul in fun, recommendation on February 23, 2009
Damn you, Mr. Bezos!
Posted by Paul in fun, Music, recommendation on February 22, 2009
Once again, I was blind-sided by the Amazon recommender. I was placing an order for a few books that my wife wanted. Easy enough, and it would only take 5 minutes. But while I was adding Marie’s books to the shopping cart, a recommendation for a new Keith Emerson CD caught my eye. The last thing I bought by KE was not so good, but the reviews for this CD were rather positive – and so I added it to the cart. And then another Keith Emerson Anthology CD was recommened “just for me” – which has some songs I haven’t listened to for years and are still sitting on vinyl in my attic. That 2 CD set found its way into my shopping cart too.
And then, while at the final checkout, the new Ben Fry / Processing book was sitting there, with 13 excellent reviews. How could I pass that up? And so with an extra $80 removed from my wallet, I finally checked out of the store. Really, that should be illegal. But I’m looking forward to the new tunes and the new book.
Music Machinery 