Speech dereverberation enhancement

Khalifa, Othman Omran and Hasmad, Safiyah and Abdul Malik, Noreha (2022) Speech dereverberation enhancement. In: 12th National Technical Seminar on Unmanned System Technology, NUSYS 2020, 24-25 November 2020, Online.

	PDF (Publisher's website) - Published Version Restricted to Registered users only Download (462kB) | Request a copy
Preview	PDF (SCOPUS) - Supplemental Material Download (385kB) | Preview

Abstract

Reverberation is the process of multi-path propagation of a sound from its source to a receiver. It is a blind problem where the source signal is unknown and non-stationary as well as the acoustic channel also unknown and time-varying. Therefore, the effect of reverberation on speech causes sound distant and spectrally distorted and can also reduce intelligibility. Dereverberation is therefore an important speech enhancement. However, different methods were proposed to reduce the noise with keeping the original signal as natural as possible. In this paper, a method was proposed using Wiener filter approach to enhance the reverberated speech signals. The mechanism and the theories of the Wiener filter on how to enhance the speech is fully explained. In addition, Discrete Fourier Transformation, linear spectrum normalization, correlation algorithms are explained in detail. The techniques of Wiener filter have been implemented and the desired results of enhanced speech have been obtained successfully.

Item Type:	Conference or Workshop Item (Other)
Uncontrolled Keywords:	Speech enhancement Dereverberation Noise reduction Wiener filter
Subjects:	T Technology > T Technology (General)
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Engineering > Department of Electrical and Computer Engineering
Depositing User:	Prof. Dr Othman O. Khalifa
Date Deposited:	22 Aug 2022 11:49
Last Modified:	22 Aug 2022 16:35
URI:	http://irep.iium.edu.my/id/eprint/99530

Actions (login required)

View Item