The ITx Rutherford 2019 Programme may change without notice
The purpose of this work included the annotation of audio recordings of bird vocalizations to be used to train a machine-learning algorithm to automatically detect bird calls. In addition, this work was intended to demonstrate the ability of The Cacophony Project’s mobile phone-based ‘Bird Monitor’ for on-going monitoring of bird vocalizations.
This work is important because it forms part of The Cacophony Project’s strategy to provide a low cost and robust means of collecting bird vocalization information to help determine the effectiveness of pest control activities. The main results show that the Bird Monitor does reliably capture bird calls over an extended period and can be used to create many annotated recordings from a real situation. It is concluded that the approach of choosing the distinct call of the Morepork as an entry into the area of automatic bird call counting was valid.
I originally trained as an Electronic Engineer specialising in semiconductor fabrication. In 1993 I moved into a career in Information Technology and have been involved in a number interesting positions including modelling of hydro power systems, developing an email application for young children and using domestic power consumption to determine the well being of the elderly.
I currently lecture in the areas of Management Support Systems and Management of Information Systems in the Centre for Information Technology at Wintec. My current research interest is developing Android software for The Cacophony Project (cacophony.org.nz) to record birdsong.
The Cacophony Project