HOME INDEXING CALL FOR PAPERS JOURNAL POLICY MANUSCRIPT CURRENT ARCHIVES EDITORIAL TEAM
   
TITLE : DISCRIMINATING HUMAN WHISPERS FROM PEST SOUND DURING RECORDINGS IN COCONUT PALM GROOVES USING MFCC AND VECTOR QUANTIZATION  
AUTHORS : Betty Martin      Vimala Juliet            
DOI : http://dx.doi.org/10.18000/ijabeg.10067  
ABSTRACT :

The key pest of horticultural and ornamental palm species in Asia are endangering landscape. The pest specially the Red Palm Weevil bores deep into palm crowns ,trunks and offshoots and remains hidden from visual inspection until the palms are nearly dead. Acoustic signals of this boring pest can be recorded from the infested palms using the off-the-shelf recording devices. Whispers from human during recording may result in wrong interpretations confirming the presence of some acoustic activity of the pest. This can be differentiated by using MFCC and VQ. The MFCC is applied on input for sound identification and VQ a quantization method deals with vectors. This paper illustrates how human whispers can be differentiated from pest’s acoustic activity during recording.

Keywords: Mel frequency coefficient characteristics (MFCC), Vector Quantization (VQ), Coconut Palm grooves.

 
  Download Full Paper
 
Copyrights ©Sathyabama Institute of Science and Technology (Deemed to be University).
Powered By: Infospace Technologies