Noise Reduction for Speech Enhancement Using Non-Linear Wavelet Processing

Abstract

The problem of speech enhancement presents many obstacles in the speech processing field. This thesis develops several speech de-noising systems that can be used in the time, fourier, and wavelet domains. We present two thresholding techniques: soft and hard. The application of these thresholding techniques to noisy speech data is discussed. The combination of both wavelets and the Fourier domains with noisy phase restoration proves to yield the best results in terms of intelligibility. Informal listening tests were conducted in order to compare the effects and differences between the speech de-noising systems.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 1994
Accession Number
ADA280469

Entities

People

  • Hassan Dehmani

Organizations

  • Air Force Institute of Technology

Tags

DTIC Thesaurus Topics

  • Air Force
  • Algorithms
  • Band Spectra
  • Complex Numbers
  • Complex Variables
  • Computer Programs
  • Distortion
  • Electrical Engineering
  • Fourier Analysis
  • Frequency
  • Frequency Bands
  • Frequency Domain
  • Noise
  • Noise Reduction
  • Random Variables
  • Repetition Rate
  • Signal Processing

Readers

  • Computer Vision.
  • Speech Processing/Speech Recognition.