Biography

Ben (he/him) graduated from the University of York with an MEng in Electronic Engineering with Music Technology Systems (with a year in industry). During his degree, he developed an interest in signal processing and machine learning methods for audio. His year in industry and final year project focussed on embedded programming for DSP and control systems at NST Audio. After graduating, he worked for the start-up Bonza Music as a Machine Learning engineer, developing a tool to extrapolate room impulse responses for their immersive audio application.

Currently, Ben is a PhD student supervised by Karolina Prawda and is researching the extent to which one can retrieve dynamic room geometry information from reverberant speech. Though information retrieval of room geometry has been investigated thoroughly with room impulse responses, the use of reverberant speech as a continuous signal and carrier opens up the research to applications in audio forensics, digital twinning, and robot navigation. Additionally, due to the continuous and non-invasive quality of using reverberant speech, it allows for analysis of room geometry information and how it changes as a function of time. This PhD project is in collaboration and supported by the industrial partner Treble Technologies.

Highlighted Publications

Perceptual Evaluation of Extrapolated Spatial Room Impulse Responses From a Mono Source
Ben Heritage, Fiona Ryder, Michael McLoughlin, Karolina Prawda

Contact

ben.heritage@york.ac.uk