Music: Bridging the Semantic Gap

January 9, 2006 Blog No Comments

If you’re like most of us, you’ve likely heard a song you like and thought “I wish I could find more music like that!” Researchers working on the Semantic Interaction with Music Audio Contents (SIMAC) project, are developing a prototype system that will be able to recommend music you might like based on its actual properties.

SIMAC researchers have developed a method of generating semantic descriptors (tags) for music that describe its characteristics. Songs can now be categorized by their musical properties—rhythm, beats per minute, timbre, harmony, key, structure and instrumentation—not just by artist, title, and genre.

Xavier Serra, SIMAC project manager, told Science Daily, “The technique represents a major advancement over the existing methods used by audio software. It improves the way users can organize, navigate and visualize audio files and how they can interact with music on their audio player, PC, or on the internet.”

According to the SIMAC website, “SIMAC is about music metadata, about what you can say of a piece of music, on what is hidden in a music file, in a collection of music files, and in the collective knowledge of a community of music lovers. Music audio content can be automatically extracted with the SIMAC tools in order to open up new navigation and retrieval strategies, or in order to get suggestions for discovering potentially interesting (but unknown!) music. SIMAC will make possible to step beyond music information retrieval and move towards the realm of music content discovery.”

SIMAC also created an innovative system that can recommend music to users called FoaFing the Music. It’s based on the “friend of a friend” concept and draws information about music from its semantic markup and from the user’s online profile, past purchase history, and what has been written about music by others.

U.S. software maker mSoft has reportedly licensed SIMAC technolgoy and plans to use it to search library music files to find authorless tracks and sound effects that cannot be categorized using traditional methods. Philips is reportedly planning to use the SIMAC audio analysis technology in a new MP3 player its developing.