Denoising Speech Signals by Wavelet Transform
Slavy Georgiev Mihov, Ratcho M. Ivanov, Angel Nikolaev Popov · 2009
This paper investigates the use of wavelet transform for denoising speech signals contaminated with common noises. Shown are the basic principles of wavelet transform as an alternative to the Fourier transform. The practical results obtained are based on processing a large dedicated database of reference speech signals contaminated with various noises in several SNRs. This research tends to be an extension to the practical research for speech signal enhancement for the purposes of hearing-aid devices. Here is presented an investigation of the use of wavelet theory for practical signal denoising. Studied are the potentials of wavelet transform for improving the hearing perception of humans to noise contaminated speech records. This study is a continuation of the research for speech enhancement for the needs of small portable devices and particularly hearing-aid devices (1). Fourier transform based spectral analysis is the dominant analytical tool for frequency domain analysis. However, Fourier transform cannot provide any information of the spectrum changes with respect to time. Fourier transform assumes the signal is stationary, but speech signal is always non-stationary. To overcome this deficiency, a modified method-short time Fourier transform allows to represent the signal in both time and frequency domain through time windowing function. The window length determines a constant time and frequency resolution. Thus, a shorter time windowing is used in order to capture the transient behavior of a signal; we sacrifice the frequency resolution. The nature of the real speech signals is nonperiodic and transient; such signals cannot easily be analyzed by conventional transforms. So, an alternative mathematical tool - wavelet transform must be selected to extract the relevant time-amplitude information from a signal. In the meantime, we can improve the signal to noise ratio based on prior knowledge of the signal characteristics.