Significance of Phase in DNN based speech enhancement algorithms

Most of the speech enhancement algorithms rely on estimating the magnitude spectrum of the clean speech signal from that of the noisy speech signal using either spectral regression or spectral masking. Because of difficulty in processing the phase of the short time Fourier transform (STFT), noisy phase is reused while synthesizing the waveform from the enhanced magnitude spectrum. In order to demonstrate the significance of phase in speech enhancement, we compare the phase obtained from different reconstruction methods, like Griffin and Lim, minimum phase, with that of the gold phase (clean phase). In this work, spectral magnitude mask (SMM) is estimated using deep neural networks to enhance the magnitude spectrum of the speech signal. The experimental results showed that gold phase outperforms the phase reconstruction methods in all the objective measures, illustrating the significance of enhancing the noisy phase in speech enhancement.

I. Introduction

Increase in demand for voice based assistant systems like Ok-Google and Alexa along with sophisticated deep learning techniques, lead people to work more on speech enhancement. Speech enhancement is a methodology that tries to improve speech quality and intelligibility either by enhancing speech or by suppressing noise from noise corrupted speech signal. Most of the signal processing methods try to find the characteristics of noise content and suppress them. On the other hand, most of the deep learning techniques try to learn characteristics of speech content and retrieve them.