Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/59064
Title: Gender classification using audio features in mobile devices
Authors: Tan, Ivan Yun Feng
Keywords: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Pattern recognition
Issue Date: 2014
Abstract: This FYP project is to develop a voice based gender classification mobile application on the android platform. The purpose of the project was to understand the process and techniques involved in a voice verification system. Then from the knowledge obtained, apply it into developing a mobile application on one of the most commonly found mobile platform in the current market. The application is targeted to work on phone running android version between 2.2 to 4.3. A Samsung galaxy S2 phone running version 4.1.2 is use through the development and testing period. In this application, it allows the user to generate the voice characteristic of him/her and use it to run the gender classification test swiftly and come up with the result of the possible gender. From the experiment conducted, it is concluded that increase in sample size or mixture alone will not get the optimum accuracy. Instead pairing of adequate mixture based on the sample data size is needed. Also the application had 75.88% classification accuracy. Recommendations are made on the improving the visual aspect of the interface and increasing the number of voice related functionality such as age group classification and even speaker identification. Likewise more tests from different sources can be conducted.
URI: http://hdl.handle.net/10356/59064
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
Amended FYP Report - Tan Yun Feng Ivan_U1122614H.pdf
  Restricted Access
1.54 MBAdobe PDFView/Open

Page view(s)

169
checked on Sep 26, 2020

Download(s)

13
checked on Sep 26, 2020

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.