B-CHIRP: A machine learning tool to assist education of novice birders and signal processing programmers

This paper reports on development of a bird call recognition application and signal processing coding framework referred to as the “Bird Call Heuristic-based Identification and Recognition Program” (B-CHIRP for short). The aim of this project was development of an application that serves two purposes: provides a system to assist novice birders in the identification of birds based on their sounds, and a second objective of providing a coding application framework appropriate for use teaching signal processing techniques with the added twist of bringing in a complex real-world application problem and an environmentally conscious flavor to promote nature appreciation with a systems thinking approach to engineering education. This paper focuses on the developmental aspects of the B-CHIRP framework, its performance and bird identification prediction accuracy tested using a selection of indigenous South African birds. The dimension of using the framework in an educational perspective is limited to reflections on anticipated influences and potential stumbling blocks of using the framework in a signal processing teaching contexts.