Noise Suppressor using Zero Phase Signal and Accumulated Spectrum Technique

This paper proposes a wide-band noise reduction method using temporal accumulated spectrum and zero phase (ZP) signal. In the previous study, we replace the ZP signal around the origin with the ZP signal in the second or latter period to get an estimated speech ZP signal. For very low SNR environment, reliable period estimation is difficult. This paper presents a study of period estimation in noisy speech based on temporal-spectral representation. Since harmonic structure of speech changes much more slowly than noise spectrum, spectral peaks related to period harmonics would stand out over the noise through the accumulation. In this paper, we combine robust period estimation technique and ZP noise suppressor. The major advantages of this method are that it can reduce not only stationary wide-band noises but also non-stationary wide-band noises and does not require a prior estimation of the noise spectral amplitude. Simulation results show that the proposed noise reduction method further improves the SNR especially in a low SNR environment.