Pitch estimation in noisy speech based on temporal accumulation of spectrum peaks

In this paper, we present a study on robust pitch estimation by integrating spectral and temporal information in speech. Spectrum harmonics are important representations of the speech fundamental frequency. Harmonic-related spectral peaks of speech evolve much more slowly than the spectral peaks of noise. This motivates the proposition of temporally accumulated peak spectrum (TAPS), which is computed by cumulating spectrum peaks over consecutive analysis frames. In the TAPS, harmonic-related peaks are concentrated around the fundamental frequency and its multiples, while the peaks caused by noise are irregularly distributed with relatively small amplitude. A pitch estimation method is derived based on TAPS. The peak locations on the autocorrelation of TAPS indicate the frequency separations between the harmonic peaks, which are used to estimate the fundamental frequency. The proposed method is evaluated on speech signals corrupted by white noise, speech noise and babble noise. The results of pitch estimation show that our method performs more robustly and reliably than conventional time-domain and cepstrum-domain methods.