Record Details
 
  « New Search    
   
 
Bibliographic Data
Control NumberUPD-00018197491
Date and Time of Latest Transaction20090717042852.0
Cataloging SourceDENGII
Language Codeeng
Local Call NumberLG 993.5 2008 E64 C75
Main Entry - Personal NameCristobal, Elaine Rose N.
Title StatementParameter determination for efficient vowel mapping of English loan words to the Filipino language / Elaine Rose N. Cristobal (B.S. CoE), Katherine H. Rara (B.S. ECE)
Physical Description89 leaves : ill. + 1 computer laser optical disc (4 3/4 in.)
Added Entry - Personal NameRara, Katherine H.
Collection CategoryFI
 Thesis
Summary, Etc.The filipino language borrows words from other languages such as English. These words are called loan words and they are borrowed with little or no change to the spelling and pronunciation. In determining the spelling of the load words, it is necessary to have a standard. Spelling standards associated with the consonants of the Filipino language have already been established. However, no such standard exists for the vowels. The objective of this project is to be able to provide scientific aid in creating the spelling standard for loan word vowels. Particularly, we aim to determine the parameters or features that will be used to create a mapping between the 20 English vowels found in the TIMIT Speech Corpus and the 5 Filipino vowels found in the Filipino Speech Corpus (FSC). Preliminary analysis was done on the English vowels found in the TIMIT database as well as on the Filipino vowels of the FSC. This allowed an initial inspection on the similarities of the vowels. Then, two sets recordings of the loan words were done. Energy-based vowel detection allowed automated delineation of the vowels. Feature extraction was done on the vowels of the load words. Statistical methods were then used on the features to create a vowel mapping between the loan words vowels and the Filipino vowels. Two features of speech were found to be viable in creating the vowel mapping. One is the formants which are the resonating frequencies of the human vocal tract. The first two formants F1 and F2 were used. Another feature is the Mel frequency cepstral coefficients (MFCC) which takes into consideration the processes of human hearing. A vowel mapping was created using the two features F1-F2 and MFCC. This mapping will be used by the Sentro ng Wikang Filipino to aid them in formulating the spelling standard for English loan words
LocationUP DENG-II LG 993.5 2008 E64 C75 E200000439G Room-Use Only 2008 E2-380TG
Textual Physical Form DesignatorThesis
 
     
 
Physical Location
University of the Philippines
Diliman: College of Engineering Library IILG 993.5 2008 / E64 C75
 
     
 
Digital Copy
Not Available
 
     
 
         
         
Online Catalog
Basic Search
Advanced Search
Browse Subjects
Book Cart
 
         

Text Size:
S  -  M  -  L
Copyright © 2004-2026. Philippine eLib Project
Host: U.P. Diliman University Library