Briggs, Forrest; Huang, Yonghong; Raich, Raviv; Eftaxias, Konstantinos; Lei, Zhong; Cukierski, William; Hadley, Sarah F.; Hadley, Adam; Betts, Matthew; Fern, Xiaoli Z.; Irvine, Jed; Neal, Lawrence; Thomas, Anil; Fodor, Gabor; Tsoumakas, Grigorios; Ng, Hong W.; Nguyen, Thi N. T.; Huttunen, Heikki; Ruusuvuori, Pekka; Manninen, Tapio; Diment, Aleksandr; Virtanen, Tuomas; Marzat, Julien; Defretin, Joseph; Callender, Dave; Hurlhurt, Chris; Larrey, Ken; Milakov, Maxim. 2013. The 9th Annual MLSP Competition: New Methods for Acoustic Classification of Multiple Simultaneous Bird Species in a Noisy Environment. In: Sanei, Saeid; Smaragdis, Paris; Nandi, Asoke; Ho, Anthony T. S.; Larsen, Jan, eds. Proceedings of the workshop on Machine Learning for Signal Processing; September 22-25, 2013. 8 p.
Birds have been widely used as biological indicators for ecological
research. They respond quickly to environmental
changes and can be used to infer about other organisms (e.g.,
insects they feed on). Traditional methods for collecting data
about birds involves costly human effort. A promising alternative
is acoustic monitoring. There are many advantages
to recording audio of birds compared to human surveys, including
increased temporal and spatial resolution and extent,
applicability in remote sites, reduced observer bias, and potentially
lower cost. However, it is an open problem for signal
processing and machine learning to reliably identify bird
sounds in real-world audio data collected in an acoustic monitoring
scenario. Some of the major challenges include multiple
simultaneously vocalizing birds, other sources of non-bird
sound (e.g., buzzing insects), and background noise like wind,
rain, and motor vehicles.
The 9th annual MLSP competition presented a real-world
dataset of bird sounds collected in field conditions. The goal
of the challenge was to do develop a classifier which predicts
the set of bird species present in a given ten-second audio
recording. The competition was hosted on Kaggle.com, a
platform for data mining competitions. Participation in this
competition was quite extensive; 79 teams participated, and
8 out of the 10 top-ranking teams submitted a two-page summary
of their proposed methods. This paper summarizes the
results of the competition, and highlights the ideas from those
summaries.