Music Mashups
There are loads of music pages, services and apps out there on the web and internets are using them to do all sorts of cool things. A recent snapshot of Programmable Web showed over 300 music mashups, including 40 music search apps.
Yet almost all of these mash-ups start by asking you to type in the name of a song or artist.
This sucks for new bands – I don’t know their names so how can I discover them?!
That’s why Herd It has been designed with new, undiscovered artists in mind.
Maybe you want “chill electro with a female Chilean singer” – you should check out “Juna“.
Or you’re looking for “San Diego punk rock with vocal harmonies and a killer lead guitar” – listen to “SO3“.
Herd It makes discoveries like this possible for all music. Bands can upload their music into Herd It and fans can discover by describing what they want to hear.
Sign up for Herd It Music Discovery!
Want to include YOUR BAND’s music in Herd It? Let us know!
Herd It music discovery
Last month, I posted about different types of music discovery tools available online. Now Herd It has just invented a new way to find music!
Search by description – Herd It’s music discovery engine uses results from games played by YOU and the whole herd to train machines that can listen to, understand and describe music. For example, when Herd It players say that a song is “romantic”, our listening algorithms can hear features in the music and learn to identify new romantic songs. With your help, these musical machines are learning to understand all types of music – the more you play Herd It, the smarter they get!
Herd It music discovery engine lets you find music by describing the type of songs you want. Now you can find “romantic jazz with saxophone” or “funk with a horn section to play at a party”, without knowing the names of the bands or songs.
To request an invite to be a tester of Herd It music discovery, sign up here.
Herd It color search!
Does “Paranoid” by Black Sabbath seem like a red song?
Is Oasis’s “Don’t Look Back in Anger” more of a green groove?
“Stupid Blues” by Junior Brown is almost certainly blue…
One of the minigames in Herd It asks the herd what color they think a song is. We weren’t really sure if people would consistently associate words with colors – not everyone has synesthesia or plays Herd It while tripping! So we’ve taken this data and created the super-fabulous Herd It COLOR search!
Check it out and see if you think that the colors of the rainbow match the music. Maybe you can also find out who stole me lucky charms…
Music discovery in mixed up, muddled up, shook up world
As a game, Herd It’s purpose is to connect fans, share music and make fun! But Herd It is also part of a bigger picture that’s making it possible for everyone to find the perfect song for every occasion.
When you play Herd It and tag songs with words, we learn from you how to find songs with similar descriptions. Google makes it possible to search billions of web pages and, even though you don’t know the exact page or URL you’re looking for, describing what you want with words gives pretty good results. In the same way, Herd It powers a music discovery engine that makes it possible to use words to describe the music you want, search through millions of songs (over 10 million on iTunes and now add MySpace…) and find the perfect tune!
Stay tuned – Herd It search will soon be entering a private Beta testing release. In the meantime, there are other cool, interesting and odd ways of finding the music you want though. Here are just a few that we like:
Search by similarity – “I like this song, give me more like it”. Examples: pandora.com, iTunes Genius feature.
Search by emotion – “I want a song that is dark and calm”. Examples: musicovery.com, AMG Tapestry
Search by genre - “I like hip-hop, yo!”. Examples: AllMusic.com, any record store (if you remember what they are)
Search by color - “Gimme pink songs”. Example: guitarati.com (Herd It’s color minigame will find out if this method makes any sense!)
Search by song/artist name – “I want to hear the new song by Franz Ferdinand”. Example: iTunes
Search by singing/humming – “Naahh nah nah nanana naahh, nanana naaah, hey Jude”. Examples: Shazam, MiDoMi
What other music discovery methods do you like / use?
Crowdsourcing to map the world
Herd It is a “crowdsourcing” application. We’re collecting information about music by asking lots of music fans – like you – to contribute a little bit of information. By adding this information up over the entire Herd, we can start to really understand all the world’s music. In return for contributing data to Herd It, you get to hear great music and play fun games!
There are lots of other crowdsourcing ideas popping up all over the place. Some great examples are:
GalaxyZoo.org – help scientists discover and label all the millions and billions of galaxies in the universe.
Waze.com – build maps, traffic reports and turn-by-turn directions for everywhere in the world (iPhone app is only for Israel and US so far).
reCAPTCHA – when web sites ask you to type in words to prove that you are human, you’re also helping transcribe old books.
New Music in Herd It?
Herd It + Facebook
Herd It is a Facebook application. Facebook is a great platform for our game since:
- Music is social – people like to listen, dance, sing and talk about music together
- Facebook is a great way to access a large audience so that Herd It can collect lots of data
- Casual games, quizzes and fun music apps are very popular on Facebook
Over the past year of developing Herd It on FB, we’ve learned a lot about the Facebook platform, how it works and where it strengths and weaknesses are. As time goes on, I’ll post more info so that you can learn from our experiences and develop your own Facebook application.
If you aren’t on Facebook (where have you been?!) and still want to play Herd It, let us know. We’re working on a standalone version of the game.
Herd It’s blog begins
At UC San Diego’s Computer Audition Laboratory, we work on using machines to automatically analyze and understand music. Ph.D students, undergrads, professors and musicians all collaborate and use signal processing, machine learning, data mining and, now, human computation to discover everything we can about music.
In order for a computer to understand the difference between punk and metal or happy and sad music, the machine needs to learn about how each of these types of music sound. That means that we – human experts and music fans – need to teach the computers to understand these differences. Just like a child learns to tell opera from Obie Trice, we teach our computer audition algorithms to learn to recognize all types of music.
Herd It is the way that we collect data to train the machines. When you play Herd It, apart from listening to great music, playing fun games and connecting with other music fans, you are contributing examples of words that describe music. We use these tags as examples to train computer audition system so that it can listen to, understand and, ultimately, help us discover and enjoy every song ever recorded.
So, now that you know all about Herd It, go play!
www.herdit.org
