Date of Award

2005

Publication Type

Master Thesis

Degree Name

M.A.Sc.

Department

Electrical and Computer Engineering

Keywords

Engineering, Electronics and Electrical.

Rights

info:eu-repo/semantics/openAccess

Abstract

This thesis presents a novel subband noise reduction technique for speech enhancement, termed as Adaptive Subband Wiener Filtering (ASWF), based on a critical-band gammatone filterbank. The ASWF is derived from a generalized Subband Wiener Filtering (SWF) equation and reduces noises according to the estimated signal-to-noise ratio (SNR) in each auditory channel and in each time frame. The design of a subband noise estimator, suitable for some real-life noise environments, is also presented. This denoising technique would be beneficial for some auditory-based speech and audio applications, e.g. to enhance the robustness of sound processing in cochlear implants. Comprehensive objective and subjective tests demonstrated the proposed technique is effective to improve the perceptual quality of enhanced speeches. This technique offers a time-domain noise reduction scheme using a linear filterbank structure and can be combined with other filterbank algorithms (such as for speech recognition and coding) as a front-end processing step immediately after the analysis filterbank, to increase the robustness of the respective application.Dept. of Electrical and Computer Engineering. Paper copy at Leddy Library: Theses & Major Papers - Basement, West Bldg. / Call Number: Thesis2005 .G85. Source: Masters Abstracts International, Volume: 44-03, page: 1452. Thesis (M.A.Sc.)--University of Windsor (Canada), 2005.

Download

Share

COinS