Scientists build musical search engine
Internet users will be able to source nagging tunes or half-remembered songs in the near future by simply humming to their PC, scientists have claimed.
Researchers at Queen Mary, University of London, said on Thursday they had developed an on-line music recognition system that could become the musical equivalent of the popular search engine Google.
The system, which had its first public demonstration on Thursday in Paris, currently enables users to find pieces from a selection of 3,000 classical tunes by entering a sound file into it.
The OMRAS (On-line Music Recognition and Searching) system lets Web surfers find a piece of music in a polyphonic, symbolic collection using polyphonic audio as a query. Polyphonic means many sounds at once and, according to the researchers, allows people to retrieve complex recordings such as full orchestra rather than the sound made by a single instrument or voice.
According to a spokesperson for the university, possible uses of the technology include in musical schools as a reference tool and to encourage children to interact with and take-up music.
However, while copyright laws currently restrict what can be done with the system, the research team behind the project have mapped-out some intriguing possibilities for future versions of such technology.
The team, which consisted of British and American scientists, said that computers in the future will be able to "listen" to a piece of music and then "write" it down (i.e. produce sheet music). In addition to printing out a score, it will also be possible to synthesise the sheet music onto a computer, they predicted.
The system will also enable the retrieval of similar pieces of music, as well as original or alternative recordings. For example, if you played the system one version of the song "My Way," it would find all the other versions of it.
Another prediction was that it will able to make comparisons between original music and music recorded by songwriters or DJs who lift unauthorised samples from other artists' tracks without their permission.
"Systems like ours will change the way we use music in our everyday lives," said Professor Mark Sandler of the electronic engineering department at Queen Mary, University of London. "Being able to get at the underlying musical structure, and then querying on-line search engines will make music appreciation a much more interactive experience. And for the music industry, it offers new ways to make money."
However, the university spokesperson could not detail when such future versions will become a reality or how the inevitable copyright difficulties will be overcome.