(Publisher of Peer Reviewed Open Access Journals)
ICETTR-2013
Full-Text PDF
Paper Title : Identification & Detection System for Animals from their Vocalization
Author Name : A. D. Mane, Rashmi R. A., S. L. Tade
Abstract : Until now little research has been done in the area of animal sound identification. Animal sound classification and retrieval is very helpful for bioacoustics and audio retrieval applications. This paper is a literature review of an animal identification and detection system based on animal voice pattern recognition. The system uses the Zero-Cross-Rate (ZCR), Mel-Frequency Cepstral Coefficients (MFCC) and Dynamic Time Warping (DTW) joint algorithms as the tools for recognizing the voice of the particular animal. ZCR is used for end point detection of input voice such that the silence voice can be removed. MFCCs are used as it is the most widely used features for speech recognition due to their ability to represent the speech spectrum in a compact form. DWT is used for voice pattern classification by getting the optimal path between input voice and referenced voice in database. The results are encouraging and motivate further research in this domain, particularly detecting the state of the animal like normal, hunger, sleep, heat etc.
Keywords : Zero-Cross-Rate (ZCR), Mel-Frequency Cepstral Coefficients (MFCC), Dynamic Time Warping (DTW), spectrum.
Cite this article : A. D. Mane, Rashmi R. A., S. L. Tade " Identification & Detection System for Animals from their Vocalization " ,ICETTR-2013 ,Page No : 347-352.