Using computers for a task as subjective as discerning your music tastes sounds like a tall order. Nevertheless, some commercial companies and research organizations are developing software designed to analyze the music you like and recommend tunes, even from groups you’ve never heard of.
Gracenote, a digital entertainment company, says that it will offer a product for online music stores by midyear that will help the stores make smarter music recommendations for their customers. And a project partially funded by the European Union is ready to license similar technologies to service providers and consumer electronics makers.
Early efforts at recommending music relied on signal processing techniques to uncover similarities in music, such as sound quality, according to Xavier Serra, who is managing the E.U.-funded Semantic Interaction with Music Audio Contents (SIMAC) project at Barcelona’s Pompeu Fabra University. That was enough to group tunes with the same acoustic properties, but the method might still have linked a fast-paced classical overture with a thumping techno beat.
The next generation of the technology will combine signal processing with information from databases containing input from music experts, or even personal data supplied by a consumer.