Birds are one of nature's joys, and many sections of the New Zealand community engage in some form of bird watching, from feeding garden birds to visiting our island safe havens. Birders would like to have an easy to access tool for identifying the birds that they hear. The rise of smartphone technology has enabled apps that perform this task for music, and the appeal of such an app for birdsong is clear. We focus to develop an app that:
- Does not require any further information about the bird such as images, size, colour, or where the bird is, but simply an instant recording of the bird made by the phone itself.
- Is more practical when the preferred species is nocturnal or hard to see.
- Is ready to use - does not require tuning/ training.
What New Zealand conservation problem are we trying to solve?
Management of birds is essential to ensure their survival. This app helps to
Increase public participation in bird monitoring
Provide a learning platform
Decide presence/absence of a species in a forest
Understand the overall health of a forest
Build a recogniser for processing auto-recordings
What makes our idea new and unique?
Recently we developed a successful noise reduction method for natural bird recordings that usually carry diverse noise. Noise is one of the biggest challenges in automated bird recognition. The proposed app will use our noise reduction technique as well as machine learning techniques to accomplish the recognition.