Dominant harmonic modification with data adaptive filter based algorithm for robust pitch estimation

This paper presents an efficient pitch estimation algorithm of noisy speech signal using the combination of dominant harmonic modification (DHM) and data adaptive time domain filtering approach. The noisy speech signal is pre- filtered within the range of fundamental frequency. The dominant harmonic (DH) is determined in pre-filtered signal and enhanced the amplitude of DH. Normalized autocorrelation function (NACF) is computed from the modified signal. Empirical mode decomposition (EMD) is applied to the obtained autocorrelation function to implement the data adaptive filtering. Such time domain filtering reduces the effects of the harmonics other than DH in the NACF domain. The signal obtained by the partial reconstruction in EMD domain is effectively used to determine the pitch period. The experimental results show that the proposed one performs better for noisy and clean speech signals.