Date of Award

2007

Degree Type

Thesis

Degree Name

Master of Engineering Science

Program

Electrical and Computer Engineering

Abstract

The advent and widespread use of mobile communication devices have signifi cantly increased the need for reliable speech communication in acoustically diverse environments. The speech communication applications involving digital hearing aids, mobile phones, hands-free telephones, and speech recognition systems are often used in noisy outdoor environments such as noisy streets, restaurants, subway stations, air ports, or in moving vehicles. Since the quality and intelligibility of speech drastically degrades in the presence of background noise, speecli enhancement systems form a vital front-end to improve the performance of these modern communication devices in noisy environments. This thesis contributes to the development of novel methods of subband adaptive filtering applied to a dual channel speech enhancement system. A new subband Crosstalk Resistant Adaptive Noise Canceller structure based on an improved fixed step-size affine projection algorithm is proposed (FSS-Sub-CRANC). This research work is further extended to demonstrate the advantage of employing variable step size adaptive filters over the fixed step-size adaptive filters using a variable step-size Sub-CRANC (VSS-Sub-CRANC). The proposed Sub-CRANC structures consist of a built-in Voice Activity Detector (VAD) which sequentially detects the speecli activity in each subband and two other adaptive filters connected in a feedback loop to cancel the background noise and crosstalk. An objective measurement technique called the Perceptual Evaluation of Speech Quality (PESQ) Mean-Opinion Score (MOS) is used to evaluate the performance of the proposed Sub-CRANC structures with existing schemes. The proposed Sub-CRANC schemes were evaluated with simulated and experimental data collected from hearing aids. Results show that the proposed architectures exhibit better performance than conventional techniques.

Share

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.