With the recent explosion in the quantity of digital audio libraries and databases, content descriptions play an important role in efficiently managing and retrieving audio files. As repetition and transformations of music structure create a unique identity of music itself, extracting such information can link low-level and higher-level descriptions of music content and provide better quality access plus a powerful way of interaction. In this talk, Dr. Bee Suan Ong will present her doctoral research work carried out at the Pompeu Fabra University in Barcelona (Spain) within the SIMAC project funded by the EU.

In particular, she will present her research on the discovery and extraction of unified high-level structural descriptions directly from polyphonic music signals. After an introduction to the main concept of structural description discovery and extraction, emphasis will be given on the application context in song cover identification and loop extraction.