| P. Srinivasan and L. H. Jamieson. Wavelets based on splines: an applications. SPIE, 2825(4):456-- 464, 1996. |
....each scale. Wavelet local maxima correspond to sharply changing signal features (at that scale) while local minima correspond to slowly varying phenomena. The wavelet coefficients are zero where the signal (at that scale) is constant, a fact that allows the use of such wavelets for speech coding [41]. The Wavelet Transform of Speech Signals Figure 2.6 shows an example of using the above wavelets for a speech signal. Recall from Chapter 1 that voiced speech is modeled as a vocal tract response excited by a glottal pulse. This pulse essentially CHAPTER 2. THE WAVELET TRANSFORM 16 0 100 200 ....
....due to the points of glottal closure are more coherent than the extrema due to reverberations. This observation will be utilized throughout this paper. Wavelet Extrema The extrema of the wavelet coefficients at each scale have been used extensively for signal de noising, detection, and coding [26, 41]. In fact, it has been shown that a signal f(t) can be closely approximated by reconstruction from only the lowest resolution approximation and the largest wavelet extrema of each detail signal. Mallat proposed an algorithm in [26] that numerically reconstructs an estimate, h(t) of the signal, ....
P. Srinivasan and L. H. Jamieson. Wavelets based on splines: an applications. SPIE, 2825(4):456-- 464, 1996.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC