Date of Award

2011

Degree Type

Thesis

Degree Name

Master of Engineering Science

Program

Electrical and Computer Engineering

Supervisor

Dr. Vijay Parsa

Abstract

Improving the quality and intelligibility of speech signals continues to be an important topic in mobile communications and hearing aid applications. This thesis explored the possibilities of improving the quality of corrupted speech by cascading a log Minimum Mean Square Error (logMMSE) noise reduction system with a Harmonic Speech Enhancement (HSE) system. In HSE, an adaptive comb filter is deployed to harmonically filter the useful speech signal and suppress the noisy components to noise floor. A Bandwidth Extension (BWE) algorithm was applied to the enhanced speech for further improvements in speech quality. Performance of this algorithm combination was evaluated using objective speech quality metrics across a variety of noisy and reverberant environments. Results showed that the logMMSE and HSE combination enhanced the speech quality in any reverberant environment and in the presence of multi-talker babble. The objective improvements associated with the BWE were found to be minimal

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.