Significance of Phase in DNN based speech enhancement algorithms

Rani, P. Swetha and Andhavarapu, Sivaganesh and Kodukula, Sri Rama Murty (2020) Significance of Phase in DNN based speech enhancement algorithms. In: National Conference on Communications (NCC), 21-23 February 2020, Kharagpur, India.

Full text not available from this repository. (Request a copy)


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.

[error in script]
IITH Creators:
IITH CreatorsORCiD
Kodukula, Sri Rama Murty
Item Type: Conference or Workshop Item (Other)
Uncontrolled Keywords: Complex ideal ratio mask (CIRM), Deep learning, Long short term memory (LSTM), Phase estimation, Phase sensitive mask (PSM), SMM, Speech enhancement
Subjects: Materials Engineering > Testing and measurement
Materials Engineering > Materials engineering
Materials Engineering > Nanostructured materials, porous materials
Materials Engineering > Organic materials
Materials Engineering > Composite materials
Divisions: Department of Electrical Engineering
Depositing User: . LibTrainee 2021
Date Deposited: 24 May 2021 07:35
Last Modified: 24 May 2021 07:35
Publisher URL:
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 7774 Statistics for this ePrint Item