Robust pitch estimation using a wavelet variance analysis model

In this paper, we present a robust pitch estimation algorithm with the combination of wavelet transform (WT) and variance analysis (VA) for noisy speech. The technique uses the discrete wavelet transform to extract the wavelet parameters of the speech in the fundamental frequency band, and then performs VA to generate a variance statistical distribution function of the wavelet parameters. Finally, peak detection is applied to extract the pitch period. The simulation results show that the proposed method can give a superior accuracy relative to some of the existing methods from high to very low signal-to-noise ratio (SNR) levels in white noise, pink noise and babble noise.