METHODS FOR CLASSIFICATION OF NOCTuRNAL MIGRATORY BIRD VOCALIZATIONS USING TIME-FREQUENCY REPRESENTATIONS Patti, Anand Many species of birds in Americas vocalize during nocturnal migration ights. Acoustic detection and classi cation of the calls show potential for study of the nat- ural history of these migrant birds. In particular, information about the species' composition and number of birds involved in migration movements may be obtain- able through acoustic techniques. Other methods such as radar monitoring may have capability only to assess the number, but not the composition. Mel Frequency Cep- stral Coe cients-Gaussian Mixture Model-based methods, Mel Frequency Cepstral Coe cients-Hidden Markov Model-based methods and spectrogram correlation-based methods have been proposed to automate the recognition/classi cation of the noctur- nal ight calls. Here we investigate the impact of choice of time-frequency representa- tions on Mel Frequency Cepstral Coe cients-Hidden Markov Model-based classi er and correlation based classi er performance. We use a collection of recordings of nocturnal ight calls of several species of thrushes and other bird species with similar calls to evaluate and compare classi ers. M.S. in Electrical Engineering, July 2012e Williamson, Geoffrey A 2012-07-30 2012-07 Thesis application/pdf islandora:7458 http://hdl.handle.net/10560/2917 ECE / Electrical and Computer Engineering Illinois Institute of Technology en In Copyright http://rightsstatements.org/page/InC/1.0/ Restricted Access