Bird sounds are complex and fascinating. Can we automatically "understand" them using machine learning? I will describe my academic research into "machine listening" for bird sounds. I'll tell you why it's important, methods we use, Python libraries, open code and open data that you can use. Examples of the latest research, and a successful commercial recognition app (Warblr).
This talk is an overview aimed at Python users who are interested in machine learning, wildlife, and/or sound processing. It covers topics from the academic research project I lead, all implemented in Python, with tips throughout about libraries we use, handy bits of open code (including the winning entry from a machine learning data challenge I led last year), as well as sources of open data for audio experiments.
Topics include classifying sounds, detecting sounds, how we deployed our Python research code directly as a commercial app, and a flavour of the more cutting-edge research topics we're focussing on now. This will not be a tutorial on deep learning - instead it will show examples of how we use these methods for audio analysis.