When he was a BBC R&D-funded PhD student at the University of Surrey, Toby researched audio un-mixing, developing techniques for dealing separately with the different sounds mixed into broadcast material.
Toby’s research focused on the use of time-frequency masks to separate audio mixtures. The goal of the work was to find a way of calculating mask coefficients that maximized the removal of interferer audio whilst minimizing artefacts in the separated audio. Toby’s work led to several conference papers and BBC R&D technical notes. Toby was awarded his PhD in 2015.