website menu

C4DM Seminar: Derry Fitzgerald : Projet - Spatial Audio Separation Using Projections

For announcements of future C4DM seminars, please join our mailing list.

Date and Time
Wednesday, 13th January 2016, at 3:00pm

Top Floor Meeting Room, Informatics Teaching Laboratory, Queen Mary University of London, Mile End Road, London E1 4NS. Information on how to access the school can be found at here.

Derry Fitzgerald

Projet - Spatial Audio Separation Using Projections

This talk presents a recently proposed projection-based method for the unmixing of multichannel audio signals into their different constituent spatial objects. Where previous research has processed the original multichannel mixtures directly and has been principally focused on the use of inter-channel covariance structures, here we instead process projections of the multichannel signal on many different spatial directions. We then propose an algorithm which takes these projections as the observations, discarding dependencies between them. We further show how to recover the separated spatial objects and demonstrate the use of the technique on stereophonic music signals.

Dr Derry FitzGerald is a senior Post-Doctoral Researcher in Nimbus. He was a Stokes Lecturer in Sound Source Separation algorithms at the Audio Research Group in DIT from 2008-2013. Previous to this he worked as a post-doctoral researcher in the Dept. of Electronic Engineering at Cork Institute of Technology, having previously completed a Ph.D. and an M.A. at Dublin Institute of Technology. He has also worked as a Chemical Engineer in the pharmaceutical industry for some years. In the field of music and audio, he has worked as a sound engineer and has written scores for theatre. He has recently utilised his sound source separation technologies to create the first ever officially released stereo mixes of several songs for the Beach Boys, including 'Good Vibrations', 'Help me Rhonda' and 'I get around'. His research interests are in the areas of sound source separation, upmixing from mono to stereo or 5.1 surround sound, automatic music transcription, as well as non-negative matrix and tensor factorisations.

Return to top