A real-time bird sound recognition system using a low-cost microcontroller

Publication Type:Journal Article
Year of Publication:2019
Authors:Küc̣üktopcu, Masazade, Ünsalan, Varshney
Journal:Applied Acoustics
Volume:148
Pagination:194 - 201
Date Published:Jan-05-2019
ISSN:0003682X
Keywords:Bird call processing, Feature extraction, Mel frequency cepstrum coefficients, Real-time processing, Spectral noise gating
Abstract:

Monitoring the environment using stand-alone sensor nodes provides valuable information to researchers, practitioners and policy makers. One such problem in this area is bird activity monitoring using its sound patterns. There are several existing devices that can record bird sounds in a region. Then, the stored data is processed off-line. However, the time between the acquisition and processing stages while using such devices is generally of the order of weeks. This approach is not amenable to real-time monitoring. Therefore, there is a need for a system which is able to both monitor the environment and process data on the system itself. The most important attribute of such a system is its power consumption since it operates in the environment only by its battery. In this study, we propose such a stand-alone, energy efficient, low-level, custom-made, real-time bird call processing system concentrated on single-labeled bird calls. The system is composed of a microphone, Texas Instruments Tiva C microcontroller, and a storage unit. The proposed system enables data recording, preliminary on-board signal processing, feature extraction, classification, and data storage. In the proposed system, we simultaneously record and process data. Hence, the system not only stores the environmental sounds, it also classifies the detected birds on-site. The proposed system offers flexibility (both in hardware and software) for expansion.

URL:https://linkinghub.elsevier.com/retrieve/pii/S0003682X18304213https://api.elsevier.com/content/article/PII:S0003682X18304213?httpAccept=text/xmlhttps://api.elsevier.com/content/article/PII:S0003682X18304213?httpAccept=text/plain
DOI:10.1016/j.apacoust.2018.12.028
Short Title:Applied Acoustics
BioAcoustica ID: 
Non biological: 
Taxonomic name: 
Scratchpads developed and conceived by (alphabetical): Ed Baker, Katherine Bouton Alice Heaton Dimitris Koureas, Laurence Livermore, Dave Roberts, Simon Rycroft, Ben Scott, Vince Smith