Please use this identifier to cite or link to this item:
|Title:||Audio signal analysis||Authors:||Suxan Tanzil||Keywords:||DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Signal processing||Issue Date:||2009||Abstract:||Audio signal classification (ASC) involves extracting relevant features from a sound, where they will be used to identify into which of a set of classes the sound is most likely to fit. The feature extraction and classification algorithms used can be diverse depending on the classification domain of the application. In this project, the author first constructed a sound database containing the audio files to be classified. The sound database was created by recording the sounds from movies or downloading them from internet. The types of sounds included in the database are cough sounds, cup-platter sounds, door opening and closing sounds, and telephone ringing sounds. As mentioned above, the first step for ASC is to perform feature extraction. There are a lot of algorithms can be used for feature extraction. One of the most popular methods, which was also employed in this project, is Mel-Frequency Cepstral Coefficients (MFCC). For the classification method, this project employed the most popular one, which is Support Vector Machine (SVM). MATLAB was chosen as the tool to conduct the computer simulation. An MFCC algorithm was written in MATLAB code and OSU-SVM toolbox for MATLAB was downloaded from internet. The simulation results under different parameter values are provided in this report to show the performance of the system.||URI:||http://hdl.handle.net/10356/17855||Rights:||Nanyang Technological University||Fulltext Permission:||restricted||Fulltext Availability:||With Fulltext|
|Appears in Collections:||EEE Student Reports (FYP/IA/PA/PI)|
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.