2026-08-27T05:32:42Z https://meral.edu.mm/oai

oai:meral.edu.mm:recid/3894 2021-12-13T01:00:43Z 1582963302567:1597824273898 user-ucsy

Enhancing Speech in Noisy Background Using Spectral Subtraction Han, Wai Phyo The presence of background noise in recorded speechreduces the quality of speech. It can also degrade theperformance of other systems, such as speech compression,speech recognition, and speaker authentication. The aim ofspeech enhancement methods is to improve the quality of speechin noisy environment. The main objective of this paper is todevelop a speech enhancement system based on the spectralsubtraction. Using Fast Fourier Transform (FFT), thesemethods firstly transforms the spectra if no-isy input speech intothe magnitude signal spectra. To get the enhanced outputspeech, the expected noise spectrum is subtracted from the noisyspeech spectra. The voices of male and female are used to testas input signals in this system. The effectiveness of the algorithmis tested by speech signals which are corrupted by the stationarynoise and non-stationary noise at varying Signal-to-NoiseRatios (SNRs). The quality of the enhancement speech is testedsubjectively and objectively. 2009-08-03 http://hdl.handle.net/20.500.12678/0000003894 https://meral.edu.mm/records/3894