Application of instantaneous frequency estimation for fundamental frequency detection
Haiyun Yang, Lunji Qui, Soo-Ngee Koh · 2002
Speech signals are nonstationary processes in nature. To represent the nonstationarity, time-frequency analysis techniques are suitable tools since they represent signals simultaneously in both time and frequency. In this paper, the instantaneous frequency (IF) technique is used for fundamental frequency estimation. Since a speech signal is a multicomponent signal, the wavelet transform (WT) which functions as a special bandpass filter, is used to eliminate and attenuate the harmonics of the speech signals. The transformed speech signal is approximately a monocomponent signal. The fundamental frequency of the speech signal can be determined from the estimated IF of the transformed signal. This new method can be used to detect the true nonstationarity of the speech signal as it provides an estimate of the fundamental frequency for each sampling instant.>