Analisis Spektral Dalam Prapemprosesan Sistem Pengecaman Suara

AHMAD DARUS, MOHAMAD ZAHARUDDIN (2003) Analisis Spektral Dalam Prapemprosesan Sistem Pengecaman Suara. Other thesis, Universiti Teknologi Malaysia.

[img]
Preview
PDF
mohammadzaharuddin_ahmad_darus.pdf

Download (1828Kb) | Preview

Abstract

Speech recognition system is on the popular activity nowadays and the progress of the development is parallel to the millennium that is the interaction between human and machine. Even though the technology is well developed, there are problems that have to be changed and improved from time to time. Amongst the problem the rise are noises that caused by the sound of spinning fan, falling rain and et cetera when speech are to be put in. Therefore, we have to avoid problems like this during the speech recognition preprocessing level that is spectral analysis. In this project, the problem is solved by using the fast Fourier transform technique (FFT). FFT is used to convert the discrete or digital signal in time domain condition to discrete signal in a frequency domain condition and to act as a filter to separate the noise from speech signal. The result of this project are, the display after the change from analog to discrete signal in time domain and also the display after the conversion of speech signal to discrete signal in frequency domain.

Item Type: Thesis (Other)
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Faculty of Computer Science and Information System > Software Engineering
Depositing User: Unnamed user with email knizam@utm.my
Date Deposited: 04 Jul 2013 07:18
Last Modified: 04 Jul 2013 07:18
URI: http://ir.fsksm.utm.my/id/eprint/2613

Actions (login required)

View Item View Item