| | |
Summary: FBEM: A FILTER BANK EM ALGORITHM FOR THE JOINT OPTIMIZATION OF
FEATURES AND ACOUSTIC MODEL PARAMETERS IN BIRD CALL CLASSIFICATION
Wei Chu, Abeer Alwan
Department of Electrical Engineering,
University of California, Los Angeles,
Los Angeles, 90095
{weichu, alwan}@ee.ucla.edu
ABSTRACT
This paper extends the expectation-maximization (EM) algorithm to
estimate not only optimal acoustic model parameters, but also op-
timal center frequencies and bandwidths of the filter bank used in
cepstral feature extraction for bird call classification. The search is
done using the gradient ascent method. Filter bank and model pa-
rameters are optimized iteratively. Experiments are conducted on
a large noisy corpus containing Antbird calls from 5 species. It is
shown that features extracted using the optimized filter bank result
in a lower classification error rate than those extracted using a Mel-
scaled filter bank.
Index Terms-- EM algorithm, filter bank, bird call classification
1. INTRODUCTION
|
