Voiced/Unvoiced Detection of Speech Signals Using Empirical Mode Decomposition Model

Abstract

This paper presents a new technique for voiced/unvoiced (V/UV) discrimination based on the extraction of pitch period. Empirical mode decomposition (EMD) is employed for multi-band representation of speech signal in time domain. The fundamental oscillation in a speech segment is determined in the autocorrelation function (ACF) of the EMD space. A damped cosine model is fitted using least squared method to extract the frequency of the fundamental oscillation. The mean fractional energy contributed by the oscillations with pitch period in different ACFs is used as the factor to classify a speech segment as V/UV. The experimental results show that the performance of the proposed method is noticeable as compared to other reported methods.