|Title:||Advanced time-frequency analysis for speech processing|
|Subject:||Speech processing systems|
Hong Kong Polytechnic University -- Dissertations
Department of Electronic Engineering
|Pages:||69,  leaves : ill. ; 30 cm|
|Abstract:||The primary objective of this project is to investigate the Time-Frequency Distribution/Representation (TFD/TFR) of speech signals, using Spectrogram and Generalised Time-Frequency Representation (GTFR). The work is beneficial to applications involving short-time analysis of speech signals in the area of speech recognition, coding and synthesis. In order to visualize the speech signals in different representations, a graphic display algorithm is established. This has been accomplished through the LINUX/MOTIF environment and a toolkit called Histoscope widgets. The time-frequency represenations of speech signals derived from the TIMIT database (clean speech) have been plot and compared. By means of a sinusoidal signal, the effect of suppressing the cross-terms and frequency bandwidth broadening are highlighted. Parameters such as window advance, window length and kernel slope are adjusted to provide further information regarding the schematic process of speech signal. The results show that a cone-shaped kernel GTFR is able to overcome the tradeoff between time and frequency resolution as well as improving the accuracy of the frequency span in a spectrogram.|
Files in This Item:
|b14719691.pdf||For PolyU Staff & Students||3.95 MB||Adobe PDF||View/Open|
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item: