New music-search system would do the “listening” for you
Courtesy IST Results
and World Science staff
You like a certain song and want to hear others like it, but don’t know how to find them?
The developers of a new music-finding system say they can solve the needle-in-a-haystack problem of searching for music on the Internet, or even in your own computer.
The system, currently under development, would let songs be described not just by artist, title and genre but by their musical properties such as rhythm, timbre, harmony, structure and instrumentation.
This would provide a way to compare songs and let listeners find little-known tracks that suit their tastes but may otherwise go unnoticed, the developers say. And that would in turn, they add, level the playing field in a music industry notorious for favoring those with connections and fat wallets.
“The music world is highly commercial and only the works of the biggest artists are really well known and widely promoted,” said Xavier Serra at Barcelona’s Pompeu Fabra University. Serra manages for the project to develop the new system, dubbed the SIMAC project
( Semantic Interaction with Music Audio Contents).
“Something like 10 per cent of music accounts for 90 per cent of music sales, while the remaining 90 per cent accounts for just 10 per cent of sales. This system could therefore herald a revolution for little-known music and artists.”
Existing classification systems, such as those used to generate play lists on popular PC media players and MP3 devices, classify musical selections by artist, the track, album and genre. Users can find other tracks by the same artist, from the same album or within the same genre but there is no guarantee that the songs will be anything but remotely
alike, Serra said.
Alternatively, websites and online stores selling music often give users recommendations based on their personal preferences and the past purchases they and other customers have made.
But that system tends to cut down the options available and to favor already successful artists, Serra contends. “The ‘since you bought this artist, you might also want to buy this one, as other customers with a similar profile did’ method is not that effective,” he said, “because there may be similar songs out there but if they haven’t been bought that much they won’t appear as a recommendation.”
SIMAC “represents a major advancement over the existing methods used by audio software,” Serra said. “It improves the way users can organise, navigate and visualise audio files and how they can interact with music on their audio player, PC or the Internet.”
The system makes use
of machine learning, a type of computer program in which a machine can
steadily correct its own mistakes and improve its performance, he added.
Another component of the system is a music recommender for users to obtain recommendations that interest them about new or old songs from online stores.
It’s based on a concept called Friend of a Friend, or “FoaF,”
which draws on information from thousands of Web pages via news feeds.
Called “FoaFing the Music,” the system would use not only the musical characteristics of songs to recommend similar ones but also the users’ profile, their past purchasing history and what has been written about the songs in website news and reviews.
Serra said the system will completed by March and that some companies
have expressed interest in distributing the product, though he didn’t
say which companies.
* * *
Send us a comment
on this story, or send
it to a friend