Efficient Recognition and Improving the Performance of Automatically Classifying Audio Recordings of Bird Sound Using Machine Learning Techniques
Abstract
As an area of
interest in ecology is monitoring animal populations to better understand their
behaviour, biodiversity and population dynamics. Acoustically active birds can
be automatically based on their sounds and a particularly useful ecological
indicator is the bird, as it responds quickly to changes in its environment.
This can be done by using the method that is only for purely audio-based bird
species recognition through the application of support vector machines. The
deep residual neural network that has to be trained on one of the largest bird
song data set in the world so as to classify bird species based on their song
or sound. The existing systems on this subject has various disadvantages in
term of cost, efficiency or the maintenance of their records or the data
collected for the longer period of time. The proposed technique is followed by
extracting cepstral features on mel scale of each audio recording from the
collected standard database. Extracted mel frequency of cepstral coefficients
formed a feature matrix. This feature matrix is then trained and tested for
efficient recognition of audio events from audio test signals. Once the bird
species is identified then it is even possible to get few features regarding
that bird using this system.
Country : India
1 Pedasanaganti Swetha Nagasri
Assistant Professor, Department of Computer Science And Engineering, Malla Reddy College of Engineering for Women, Hyderabad -500100, Telangana, India
Anshul Thakur1, Pulkit Sharma2,
Vinayak Abrol3, Padmanabhan Rajan1 in “Conv-Codes: Audio Hashing For Bird
Species Classification” in Proc. of the ICASSP 2019 – 2019 IEEE International
Conference on Acoustics, Speech and Signal Processing (ICASSP),Brighton, United
Kingdom, United Kingdom, 12-17 May 2019.
Peter Jancovic, Munevver Kokuer in
“Bird Species Recognition Using Unsupervised Modeling of Individual
Vocalization Elements” IEEE/ACM Transactions on Audio, Speech and Language
Processing (TASLP), Volume 27 Issue 5,May 2019 Page 932-947.
RA. Thakur, V. Abrol, P. Sharma and
P. Rajan in “Compressed Convex Spectral Embedding For Bird Species
Classification” in Proc. of the ICASSP 2018 - 2018 IEEE International
Conference on Acoustics, Speech and Signal Processing (ICASSP), April 2018.
Md. Shamim Towhid , Md. Mijanur
Rahman in “Spectrogram Segmentation for Bird Species Classification based on
Temporal Continuity” in Proc. of the 2017 20th International Conference of
Computer and Information Technology (ICCIT)Dhaka, Bangladesh, 22-24 Dec. 2017.
