In-Ear Audio and Speech Processing Algorithms: Evaluation, Optimization and Consolidation - Year two

Everyday more than one million Canadians working in noisy environments are using Hearing Protection Devices (HPDs) to be protected against the risk of noise-induced hearing loss (NIHL). Together with the NSERC-EERS Industrial Research Chair in In-Ear Technologies (CRITIAS), the EERS Global Technologies Inc. has recently brought to market an innovative line of HPDs, enabling workers to effectively communicate in noise while being protected. More recently specialized in-ear algorithms have been developed by the researchers of CRITIAS chair: the first one is dedicated for capturing the speech, denoised and enhanced it; the second one associates with the classification of non-verbal audio events to enable health monitoring applications; the last one deals with the extraction of biosignals to enable activity tracking for workers. While the first algorithm has been already implemented successfully in EERS’s award-winning SonX product, further improvements will be made to it during this proposed Postdoctoral Fellowship. Besides, in-ear algorithms developed by CRITIAS will be implemented and optimized on a single DSP chip to ease their use by EERS. Furthermore, the development and qualification of a speech intelligibility framework dedicated for in-ear speech will be performed. This project will enable the continuous improvement of unique hearing protection technologies by EERS.

Farid Moshgelani
Superviseur universitaire: 
Jérémie Voix
Partner University: