Speeding up training of automated bird recognizers by data reduction of audio features
Allan G. de Oliveira,
Thiago M. Ventura,
Todor D. Ganchev,
Lucas N.S. Silva,
Marinêz I. Marques,
Karl-L. Schuchmann
Affiliations
Allan G. de Oliveira
Computational Bioacoustics Research Unit (CO.BRA), National Institute for Science and Technology in Wetlands (INAU), Universidade Federal de Mato Grosso, Cuiabá, Mato Grosso, Brazil
Thiago M. Ventura
Computational Bioacoustics Research Unit (CO.BRA), National Institute for Science and Technology in Wetlands (INAU), Universidade Federal de Mato Grosso, Cuiabá, Mato Grosso, Brazil
Todor D. Ganchev
Computational Bioacoustics Research Unit (CO.BRA), National Institute for Science and Technology in Wetlands (INAU), Universidade Federal de Mato Grosso, Cuiabá, Mato Grosso, Brazil
Lucas N.S. Silva
Computational Bioacoustics Research Unit (CO.BRA), National Institute for Science and Technology in Wetlands (INAU), Universidade Federal de Mato Grosso, Cuiabá, Mato Grosso, Brazil
Marinêz I. Marques
Computational Bioacoustics Research Unit (CO.BRA), National Institute for Science and Technology in Wetlands (INAU), Universidade Federal de Mato Grosso, Cuiabá, Mato Grosso, Brazil
Karl-L. Schuchmann
Computational Bioacoustics Research Unit (CO.BRA), National Institute for Science and Technology in Wetlands (INAU), Universidade Federal de Mato Grosso, Cuiabá, Mato Grosso, Brazil
Automated acoustic recognition of birds is considered an important technology in support of biodiversity monitoring and biodiversity conservation activities. These activities require processing large amounts of soundscape recordings. Typically, recordings are transformed to a number of acoustic features, and a machine learning method is used to build models and recognize the sound events of interest. The main problem is the scalability of data processing, either for developing models or for processing recordings made over long time periods. In those cases, the processing time and resources required might become prohibitive for the average user. To address this problem, we evaluated the applicability of three data reduction methods. These methods were applied to a series of acoustic feature vectors as an additional postprocessing step, which aims to reduce the computational demand during training. The experimental results obtained using Mel-frequency cepstral coefficients (MFCCs) and hidden Markov models (HMMs) support the finding that a reduction in training data by a factor of 10 does not significantly affect the recognition performance.