Deep Learning Technologies for Acoustic Echo Cancellation in Dynamic Environments

In a full-duplex hands-free voice communication system, a speaker located in a room at one end of the link may receive an echo of his/her voice due to the acoustic coupling between the loudspeaker and the microphone at the other end of the link. The goal of acoustic echo cancellation (AEC) is to remove such undesirable echo to improve the quality and intelligibility of the voice communication. While numerous AEC algorithms based on traditional adaptive filtering techniques have been proposed in the past, recent studies on the application of deep neural networks (DNN) to this problem have shown remarkable performance. However, the limited ability of these DNN to generalize to the wide dynamics of the acoustic environment still remains an open issue. The main objective of this project is to develop new DNN-based AEC algorithms to overcome this limitation. Our proposed work mainly includes incorporating additional information provided by the estimated acoustic impulse response into the DNN-based AEC framework. The new algorithms developed in this project will be used by our partner, Fluent.ai, to establish a new line of embedded voice user interfaces for communication and speech recognition, thereby enabling the company to attract additional customers and grow its business.

Faculty Supervisor:

Benoit Champagne

Student:

Hanwook Chung

Partner:

Fluent.AI Inc

Discipline:

Engineering - computer / electrical

Sector:

Professional, scientific and technical services

University:

McGill University