No abstract

Annotation of phrases in birdsongs can be helpful to behavioral and population studies. To reduce the need for manual annotation, an automated birdsong phrase classification algorithm for limited data is developed. Limited data occur because of limited recordings or the existence of rare phrases. In this paper, classification of up to 81 phrase classes of Cassin's Vireo is performed using one to five training samples per class. The algorithm involves dynamic time warping (DTW) and two passes of sparse representation (SR) classification. DTW improves the similarity between training and test phrases from the same class in the presence of individual bird differences and phrase segmentation inconsistencies. The SR classifier works by finding a sparse linear combination of training feature vectors from all classes that best approximates the test feature vector. When the class decisions from DTW and the first pass SR classification are different, SR classification is repeated using training samples from these two conflicting classes. Compared to DTW, support vector machines, and an SR classifier without DTW, the proposed classifier achieves the highest classification accuracies of 94% and 89% on manually segmented and automatically segmented phrases, respectively, from unseen Cassin's Vireo individuals, using five training samples per class.

In this paper, we present simulations and experimentally collected bird song data collected using a modified Voxnet acoustic array node (with four microphones) to perform 3D direction-of-arrival (DOA) estimation of various bird sources. We used the Approximate Maximum-Likelihood (AML) algorithm to construct the steering matrix in the beamforming process for the estimation of the DOA of the bird signals. While the computational burden is high in the 3D scenario, various strategies have been developed to reduce the computational burden of the algorithm for potential real-time applications. Extensive simulations and experimentally collected data are used to validate the effectiveness of the AML algorithm for 3D estimations and the usefulness of the modified Voxnet node. Both the estimated azimuths and elevations have approximately plus and minus 10 degrees of errors.

In this paper, we present a series of experiments on the automated classification of Cassin's Vireo individuals from song phrases using support vector machines and from sequences of song phrases using hidden Markov models. Experimental results show that accurate classification of bird individuals can be achieved using these two different levels of description of bird songs.

An automated system capable of reliably segmenting and classifying bird phrases would help analyze field recordings. Here we describe a phrase segmentation method using entropy-based change-point detection. Spectrograms of bird calls are often very sparse while the background noise is relatively white. Therefore, considering the entropy of a sliding time- frequency window on the spectrogram, the entropy dips when detecting a signal and rises back up when the signal ends. Rather than a simple threshold on the entropy to determine the beginning and end of a signal, a Bayesian recursion-based change-point detection(CPD) method is used to detect sudden changes in the entropy sequence. CPD reacts only to those statistical changes, so generates more accurate time labels and reduces the false alarm rate than conventional energy detection methods. The segmented phrases are then used for training and testing a sparse representation(SR) classifier, which performs phrase classification by a sparse linear combination of feature vectors in the training set. With only 7 training tokens for each phrase, the SR classifier achieved 84.17% accuracy on a database containing 852 phrases from Cassins Vireo (Vireo casinii ) phrases that were hand-classified into 32 types. [This work was supported by NSF.]