Voice Processing and AI from Academia to Industry¶

Pritish Chandna¶

I am Pritish Chandna, a recent graduate of the PhD program at the Music Technology Group in the Universitat Pompeu Fabra (UPF). The thesis was titled Neural Networks For Singing Voice Processing In Polyphonic Music Signals and involved source separation using synthesis for contemporary popular music and ensemble singing. I recently started working in a company called VoiceMod in Valencia, Spain, with work involving real time voice processing and conversion, This presentation includes a brief overview of the work I did during the thesis and the research I have been doing in the industry including the synergy and distinction between the two strata.

VoiceMode product intro video