Pitch detection algorithm based on normalized correlation function and central bias function

To reduce the halving and doubling errors and estimate pitch reliably even at negative signal-to-noise ratios (SNR), an algorithm based on normalized cross-correlation function and central bias function is proposed in this paper. Three pitch candidate values extracted by the half-wave rectified version of the normalized correlation function in time domain are combined with the central offset calculated by central bias function in frequency domain to detect the pitch. In order to demonstrate the efficacy of the proposed method, simulations based on Keele pitch extraction reference database are conducted at different SNR levels. A comprehensive evaluation of the pitch estimation results shows that the proposed algorithm has better robustness and precision than some of the existing methods in terms of gross pitch errors.