The Echo Nest - powering intelligent digital music applications

Meet the Midemlab 2012 Finalists

This is a guest post adapted from James Martin of midemblog about the music apps that were selected for the midemlab competition in Cannes, France next week. The Echo Nest’s Paul Lamere and Eliot Van Buskirk will speak at midem 2012, where both will also help judge the midemlab music app competition.

Alongside The Echo Nest, other companies represented on midemlab’s panel of judges include bluenove, The Coca-Cola Company, Facebook, GigaOM, Groupon, H-Farm, Mangrove Capital Partners, MOG, Music Ally, National Digital Council, Next Big Sound, Sony Music Entertainment, and SoundCloud.

Judges will examine the following music apps:

*Midemlab 2012 Finalists: Marketing, Social Engagement*

BandApp is exactly that: a tool which enables bands to make their own mobile apps. Social sharing is especially pushed [for now, the apps are down].

Oocto is a French startup whose platform allows artists to crowdfund from fans; find people to work with; and to involve fans in promotional activities.

iPluggers plugs your music to thousands of radio stations worldwide. Bands notably can track which stations have downloaded their tracks for airplay.

Crowdsurfing, a service of LiveOne Group, allows users to watch music events such as concerts online together, i.e. with their social media connections, in real time.

“On RockStar Motel, you are the record label,” says the pitch. In other words, this website offers you the possibility to represent artists and win points for getting them noticed. Superfans empowered, in a nutshell.

Tastebuds.fm is a dating site which matches people up based purely on their musical taste. Why didn’t anyone think of that before?

Beloola is a virtual world for music fans, that works in any web browser. Avatar creation, social integration and much more.

webdoc allows anyone to create rich media flyers, integrating video, photo, audio, widgets, feeds and more. No webmastering skills required.

Oleapark is a networking tool for events. It allows you to see what are the hottest tech events in your area right now, plus who is there, the most relevant people to meet, and where they are. Worth trying out in Cannes.

Pipe is a Berlin-based startup who describes its offering as “file transfer made easy” (video currently unavailable; check them out at midem).

*Midemlab 2012 Finalists: Direct-to-Consumer Sales & Content Monetization*

Cleeng is a tool for content publishers which enables them to hide parts of their content and allow access on a pay-per-view basis. No need to lock everything away in a members only premium zone of your site.

Getonic allows for the sale of content on social networks like Facebook, allowing your fans to become sales delegates. Oh and they use 2 Unlimited in this trailer, which is bound to win them some points.

Menyou – short for “me and you” – is a platform designed to enable its users to monetize content. Like Getonic, it allows anyone to resell content, via dedicated widgets, which can be posted anywhere.

Musicsoft Arts makes music apps such as DJ Mixer, of which branded – and free – versions have been made for DJs like Carl Cox (video). Their apps offer numerous monetization features, including in-app purchasing.

Mybee is a “demonetized payment system” aimed notably at music festivals, using RFID and barcode technology to pay for services like food, drink or merchandise, which are debited from a central, prepaid account.

Ondango allows brands to sell products directly on their Facebook pages, with special features like daily deals. Simple and effective.

WildChords aims to address the 85% of people who quite learning a musical instrument by making learning into a game. The player is as such rewarded for striking the right chords on the guitar… by being followed by monkeys.

sonicangel is a crowdfunding platform with a difference: it has built-in social media monitoring to keep track of which bands are the most tweeted-about right now.

Tunerights is “a stock exchange for music,” so essentially a crowdfunding platform too, as “fans become investors” in bands; the focus here is particularly on simplicity and revenue sharing.

Weezevent [no video] is an all-in-one ticketing tools that allows event organisers to create their own ticketing service, free of charge. No English-language video demo is available; but you can meet them and their fellow finalists at midem.

*Midemlab 2012 Finalists: Music Discovery, Recommendation, and Creation*

Made by Fraunhofer IDMT — co-creators of the MP3 format, no less — Song2See is a Singstar-like game for learning to play musical instruments.

MPme Built by former Shazam staffers, MPme listens to over 50,000 radio stations and matches their playlists with your tastes, notably from your own music collection, and creates sets of recommended radio stations.

ChartsNow is a mobile app which scans the charts overnight to ensure you have the biggest hits on your phone every day. It works offline, and offers one-click purchasing.

Twusic enables you to keep tabs on tracks your friends and followers are tweeting.

musicplayr works in a similar way to MP3 blog music aggregator exfm, but across multiple platforms, namely YouTube, Soundcloud and music blogs.

Rexly is an iPhone app which allows you to see in real time what your friends are listening to in iTunes. You can then listen to, rate, share and buy those songs.

WatZatSong You know how the likes of Shazam only recognises recorded music? WatZatSong claims to recognize songs you sing or hum into your computer, notably by asking its online community.

WhoSampled.com WhoSampled.com “allows music fans to explore the DNA of their favorite music,” by tracking songs over the past thousand years, no less! Direct comparisons of, say, how Kanye West sampled Daft Punk are just a click away.

Navegas [no video] is a web-based music player for all music sources, be they your MP3 collection, YouTube, SoundCloud or many others. A desktop widget system makes the whole thing easily manageable.

Using the EMI Premium Content Sandbox

The EMI and The Echo Nest have partnered to put premium content into the hands of the developer. This post shows you how you can get access to the content in the sandboxes.

Before you can get started with a sandbox, you will need to apply for access to the specific sandbox. To do so, start at The EMI Sandbox page . This page lists the available EMI sandboxes, from EMI Selection, Blue Note or one of a selection of EMI artists. Pick the sandbox that you are interested in, and read through the approval process. This is important stuff, since it outlines how you will be able to use the content. If you are not careful, you could end up spending time developing an application that you will not be able to release. Make sure to read the general EMI Sandbox FAQ.

Once you understand the application approval process, follow the directions for applying to get access to the sandbox. Once you’ve been granted access, your API key and credentials will be authorized to access content in the sandbox.

There are two types of EMI Sandboxes:

• General collection sandboxes that contain tracks, album art and previews for a wide range of artists.
• Artist specific sandboxes that contain audio, video, artwork, images, press releases and more for a specific artist.

On your Account Sandboxes page you’ll see a list of all available sandboxes, along with your approval status for each sandbox.

To get a general idea of what is in a sandbox, visit the specific sandbox by following the link fromThe EMI Sandbox Page. For example, the Blue Note page describes the available assets as ‘(1) audio for approximately 10,000 tracks and (2) Audio Metadata’

Artist-specific pages also include a developer pack which includes overview information for the artist and detailed inventory and description of all of the assets provided in the sandbox. An example is The Gorillaz developer pack.

To programmatically list the contents of a sandbox, use the sandbox/list method. This method gives a complete inventory of every asset in the sandbox. You do not need to be approved to access a sandbox to call this method, list the assets, and see what is in the sandbox. To list the contents of a particular sandbox you will need the Sandbox Key. You can find the Sandbox Key for each sandbox on your Account Sandboxes page.

For example, the EMI Blue Note sandbox has a Sandbox Key of ‘emi_bluenote’. To list the contents of this sandbox, make the method call:

    http://developer.echonest.com/api/v4/sandbox/list?api_key=N6E4NIOVYMTHNDM8J&sandbox=emi_bluenote&results=2

This call returns a response of:

    {
        "response": {
            "status": {
                "code": 0, 
                "message": "Success", 
                "version": "4.2"
            }, 
            "start": 0, 
            "total": 23604
            "assets": [
                {
                    "echonest_ids": [
                        {
                            "foreign_id": "emi_bluenote:track:EUEDD0270749", 
                            "id": "TRNVKOX13346FF118A"
                        }
                    ], 
                    "filename": "emi_bluenote/EUEDD0270749-preview.mp3", 
                    "id": "af5860829140c3a3f6ca48c29b91c481", 
                    "type": "preview_audio"
                }, 
                {
                    "echonest_ids": [
                        {
                            "foreign_id": "emi_bluenote:track:EMIDD1009985", 
                            "id": "TRKOUMM13346FF0F16"
                        }
                    ], 
                    "filename": "emi_bluenote/EMIDD1009985.mp3", 
                    "id": "8ef1b79639e4dca2c153dc3d7c98f5a5", 
                    "type": "audio"
                }
            ], 
        }
    }

This response shows that there are 23,604 assets in the sandbox. Each asset is described with at least the following information:

• type - The type of asset. Types include ‘audio’, ‘preview_audio’, ‘release_image’ and others.
• id - The asset id. Use this with the access method to gain access to the actual asset. We describe how to do this in more detail below.echonest_idsIf the asset is associated with a particular Echo Nest item such as an artist or a track, the ID will be included in this list of dictionaries.

Since the sandboxes are static, your application typically needs to get the asset inventory for a sandbox just once, and can safely cache the information locally.

To get access to the assets you use the sandbox/access method. To ensure that only authorized developers get access to the assets, the access method requires a cryptographic signature. We use Oauth 1.0 for this authentication. The authentication process can be a bit complicated, so it is highly recommended that you use a library (such as pyechonest or jEN) to handle all of the authentication details. To sign a request, you add an oauth_nonce parameter (a random string of characters that differs from call to call), a timestamp (the number of seconds since the epoch) and a signature which is generated by a applying a cryptographic has such as SHA1 to your consumer key, shared secret and the base string portion of the request. Yes, this is complicated. Such is Oauth. Use a library, and read its documentation! A guide to signing requests can be found in the Oauth 1.0 Guide. Netflix also has an excellent description of how to sign requests in their Authentication Overview.

The authentication procedure requires a Consumer Key and a Shared Secret. These can be obtained on your Account Profile page. You should avoid exposing your shared secret to the public (such as an open source repository, or in the Javascript on a web page). If a shared secret is inadvertently exposed, let us know and we will expire the secret. Here’s an example request that shows how to get access to the asset with the ID ‘af5860829140c3a3f6ca48c29b91c481’. Note, of course, that the security credentials in this example are not valid credentials. This particular example with fail with an authentication error.

    http://developer.echonest.com/api/v4/sandbox/access?api_key=N6E4NIOVYMTHNDM8J&sandbox=emi_bluenote&id=af5860829140c3a3f6ca48c29b91c481&oauth_consumer_key=b30f027ead14e3056d8833b9e783a526&oauth_nonce=1234367&oauth_timestamp=1319830594&oauth_signature_method=HMAC-SHA1&oauth_signature=tR3+Ty81lMeYAr/Fid0kMTYa/WM=

        "response": {
            "status": {
                "version": "4.2",
                "code": 0,
                "message": "Success"
            },
                "assets": [
                    {
                      "url": "http://assets-cdn.echonest.com/emi_bluenote/EUEDD03083622-preview.mp3?Expires=1320148855&Signature=HD",
                      "id": "c1b89c9b9e0ee9e53650f1d4e393d716"
                    },
                  ]
        }
    }

For each requested ID, the access method returns a dictionary containing the URL to the asset. The URL expires in 10 minutes.

Many of the assets are directly associated with Echo Nest data. For instance, an audio asset may be associated with an Echo Nest Track ID. The Echo Nest IDs for an asset are returned in the echonest_ids block of the asset description. You can use the Echo Nest API to get more detailed information about the asset. For example, one of the Blue Note tracks has an asset description like:

        {
            "echonest_ids": [
                {
                    "foreign_id": "emi_bluenote:track:EMIDD1009985",
                    "id": "TRKOUMM13346FF0F16"
                }
            ],
            "filename": "emi_bluenote/EMIDD1009985.mp3",
            "id": "8ef1b79639e4dca2c153dc3d7c98f5a5",
            "type": "audio"
        }

You can use the Echo Nest track ID TRKOUMM13346FF0F16 to get more information about the track with the track/profile call:

http://developer.echonest.com/api/v4/track/profile?api_key=N6E4NIOVYMTHNDM8J&format=json&id=TRKOUMM13346FF0F16&bucket=audio_summary

This call returns the response:

    {
        "response": {
            "status": {
                "code": 0, 
                "message": "Success", 
                "version": "4.2"
            }, 
            "track": {
                "artist": "Stan Kenton", 
                "audio_md5": "1686b479eeb8310bd65c1a9cf2708273", 
                "audio_summary": {
                    "analysis_url": "https://echonest-analysis.s3.amazonaws.com/..."
                    "danceability": 0.17244651289412521, 
                    "duration": 130.06666999999999, 
                    "energy": 0.20982863398344401, 
                    "key": 7, 
                    "loudness": -9.4399999999999995, 
                    "mode": 1, 
                    "tempo": 78.022999999999996, 
                    "time_signature": 4
                }, 
                "id": "TRKOUMM13346FF0F16", 
                "song_id": "SOUNBRE12AB0184AE3", 
                "status": "complete", 
                "title": "O Holy Night (2003 Digital Remaster)"
            }
        }
    }

This includes detailed audio information about the song including energy, danceability, loudness, key, mode and time signature. In addition, you can also get a fully detailed analysis that includes detailed timbral, pitch and loudness info, along with a fine-grained representation of the beat structure of the song.

Also included in the information is the song_id, which can be used to retrieve more cultural information about the song (hotttnesss, artist information and so on).

The general collections (EMI Selection and EMI Blue Note) are also associated with Rosetta Stone catalog allowing you to take advantage of all the APIs that support working with Rosetta catalogs. For instance, you can generate a playlist restricted to songs that are in the Blue Note catalog with a call like this:

    http://developer.echonest.com/api/v4/playlist/basic?api_key=N6E4NIOVYMTHNDM8J&artist=miles+davis&format=json&results=20&type=artist-radio&bucket=id:emi_bluenote&limit=true&bucket=tracks

Similarly, you can search for the hotttest artist in the open collection with a call:

    http://developer.echonest.com/api/v4/artist/search?api_key=N6E4NIOVYMTHNDM8J&format=json&results=20&bucket=id:emi_open_collection&limit=true&sort=hotttnesss-desc

Or you can search for the fastest songs in the Blue Note collection with a call:

http://developer.echonest.com/api/v4/song/search?api_key=N6E4NIOVYMTHNDM8J&format=json&results=20&bucket=id:emi_bluenote&limit=true&sort=tempo-desc&bucket=audio_summary

The Echo Nest API provides a vast amount of data at the Artist, Song and Track level. You can associate this data with the assets in the EMI sandbox to provide context for the content. Similarly, you can use Echo Nest searching, similarity, and playlisting APIs to help people explore for and discovery music from the EMI catalog.

The EMI Premium Content Sandboxes present a great opportunity for music application developers to create new and innovative applications around music — without needing a bevy of lawyers to negotiate label deals. We here at The Echo Nest hope you will go and build some really cool stuff with this content and our API.

The Echo Nest in The Wire (Magazine)

Our favorite music magazine, The Wire is out with a new issue featuring Echo Nest friend (and brother of a co-founder) Keith Fullerton Whitman in a huge article about his life and work. Definitely check it out, and less importantly peep The Wire’s first mention of the Nest on page 33:

Other than the nasty transposition of natural language processing to something slightly less scientific (don’t worry, we only do the former) it’s a pretty good snapshot. And congrats to KFW for the awesome article! We <3 The Wire.

The Echo Nest Powers SpotON Radio for iOS

Spotify subscribers have a great new option for listening to ad-free personalized radio on their iPhones and other iOS devices: SpotON Radio, which is already attracting a healthy degree of attention from users and the press.

Like the Spotify Radio desktop app, SpotON Radio runs on deep musical intelligence data from, you guessed it, The Echo Nest.

SpotON Radio applies The Echo Nest’s similar artists engine to Spotify’s massive catalog of over 15 million tracks, allowing users to create playlists around any artist instantly.

These stations are as easy to fine-tune as they are to create. As a song plays, users can decide whether they want the station to grow more similar to that song, or less similar. SpotON Radio then reprograms the station, allowing users to hone these stations to their specific taste. Stations are automatically saved for later listening.

Spotify attracted a lot of attention for launching an app platform within its desktop app in November, but its API for Apple iOS devices, announced in August, is another powerful tool with which developers can build Spotify apps.

We’re pleased to be powering not only the official Spotify Radio, but also SpotON Radio for iOS, which is being offered by its Swedish developer Springworks, for free, as an introductory offer.

Controlling the artist distribution in playlists

The Echo Nest engineering team just pushed out a new feature giving you more control over the artist makeup in playlists.  There is a new parameter to the playlist/static API called distribution that can be set to wandering  or focused.   When the distribution is set to wandering the artists will appear with approximately equal distribution in the playlist. If the distribution is set to focused artists that are more similar to the seed artists will appear more frequently.  When combined with the variety parameter, you have excellent control over the number and distribution of artists in a playlist.  If you want to create a playlist suitable for music discovery, create a playlist with high variety and a wandering distribution.  If you want to create a playlist that more closely mimics the radio experience choose a low variety and a focused distribution.

I’ve put together a little demo that lets you create playlists with different levels of variety and distribution settings. The demo will create a playlist given a seed artist and show you the artist distribution for the playlist.  Here’s the output of the demo with distribution set to focused:

You can see from the artist histogram that the playlist draws more from artists that are very similar to the seed artist (Weezer).  Compare to these results from a wandering playlist with the same seed and variety:

You can see that there is flatter distribution of artists in the playlist.   You can use variety and distribution to tailor playlists to the listener.  For instance, you can give the Classic Rock Radio experience to a listener by setting variety to relatively low, setting the distribution to focused and seeding with a classic rock artist like Led Zeppelin.  Here’s the artist distribution for the resulting playlist:

That looks like the artist rotation for my local classic rock radio.

Give the demo a try to see how you can use variety and distribution to match playlists to your listener’s taste.  Then read the playlist API docs to see how to use the API to start incorporating these attributes into your apps.

The Demo:  Playlist Distribution Demo (source)

(A repost from Music Machinery).  

Deadline extension for the OpenEMI App Contest

The Echo Nest Brings Twitter to Music Apps

Thanks to ongoing collaboration between Twitter and The Echo Nest, music app developers will soon be able to incorporate tweets from specific recording artists directly into their apps.

Approximately one quarter of Twitter’s verified accounts belong to musicians, many of them quite famous. The Echo Nest is applying our music intelligence platform to identify which verified accounts belong to recording artists and to deliver that data to Twitter, helping to organize Twitter verified accounts, but that’s just the tip of the iceberg. 

We are also adding these artists’ Twitter IDs to our Rosetta Stone ID mapping technology, which helps music services and app developers easily incorporate Twitter into their music apps without having to manually scour through the Twitter API to find musicians. Now, any developer can include these artists’ tweets within their apps. (Developers must comply with Twitter’s terms of service, of course.)

This move strengthens Twitter’s position as a crucial part of the internet’s infrastructure; it allows music app developers to incorporate real-time statements from popular artists into apps of all kinds; and it will help create a stronger bond between artists on their fans on a variety of platforms.

Twitter and music apps are each incredibly powerful ways for artists to reach fans — and for fans to experience artists. We’re happy to help them work together.

The first app to include tweets courtesy of The Echo Nest Rosetta Stone’s Twitter artist IDs is Discovr Music for iPhone and iPad, which adds that capability as of today. Groovebug will follow next week, after which we expect a slew of music apps to start incorporating artists’ Twitter feeds.

In the following screenshot, the Discovr Music app (iPhone version) displays Coldplay’s tweets within the app, so fans can see what that band and thousands of others are up to: 

Apps like Discovr Music can also include other fans’ mentions of Coldplay on Twitter:

You can also search for “coldplay” to find tweets about the band that don’t use the #coldplay hashtag:

Here’s what Twitter will look like in the Groovebug music app. In this screenshot, the new “it” band Foster The People’s tweets appear within the app:

Groovebug and other music apps will also feature Kanye’s latest musings:

Last but not least, our favorite Oaktown rapper to break out in 2011 is among the thousands of artists whose Twitter feeds are now in The Echo Nest’s Rosetta Stone platform:

(Follow us on Twitter: @echonest)

Video: How OpenEMI Unites Music and Technology

Here’s the old news: technology is from Mars, and music is from Venus.

Here’s the new news: They both live happily together on Planet Echo Nest.

Observe, as our own Matt Ogle and EMI’s Kara Mukerjee explain how the OpenEMI program allows app developers to build any app they can imagine, using about 12,000 well-known tracks from EMI’s legendary music catalog — Al Green, Beach Boys, Duran Duran, the Blue Note catalog, and much more — and tools provided by The Echo Nest.

The licensing is taken care of. Everybody gets paid. It’s magical. Don’t just take our word for it — “the news” agrees.

Here are Kara and Matt to explain:

Managing your API rate limit

curl -i 'http://developer.echonest.com/api/v4/artist/profile?api_key=N6E4NIOVYMTHNDM8J&name=weezer'

HTTP/1.1 200 OK
Content-Length: 135
X-RateLimit-Limit: 120
X-RateLimit-Remaining: 62
X-RateLimit-Used: 58

The Echo Nest Powers Spotify Radio

Spotify has over ten million users, 2.5 million or so of whom pay for the premium version. It also has over 15 million tracks available as free streams, and the (well-deserved) attention of the music community.

But there’s one thing Spotify didn’t have until now: artist and song radio stations. Thanks to Spotify’s deal with The Echo Nest, users of the popular music service can now create streaming radio stations based on any artist or song on Spotify. The stations are generated by The Echo Nest’s Playlist API, and are available both to free and premium Spotify subscribers.

To make this as easy as possible, Spotify populates the Create New Station menu with artists the user has already “liked.” Of course, Spotify users can also simply enter any user name to create a station that way.

The new Spotify Radio, powered by The Echo Nest, adds a powerful new discovery engine to Spotify that will not only make the service more attractive to users, but could increase their chances of converting to a paid subscription.

With over 15 million songs sitting there, it can be daunting to figure out what to listen to. Spotify Radio offers a valuable new way for music fans to find more of what they like within the music service. As they listen to Spotify Radio, they can click the currently-playing artist or track in order to explore them further. 

The new Radio app is currently available in the preview version of Spotify, and will roll out to all Spotify users over the coming weeks.