Separating overlapping bat calls with a bi-directional long short-term memory network  

作　　者：Kangkang ZHANG Tong LIU Shengjing SONG Xin ZHAO Shijun SUN Walter METZNER Jiang FENG Ying LIU 

机构地区：[1]Jilin Provincial Key Laboratory of Animal Resource Conservation and Utilization,Northeast Normal University,Changchun,China [2]School of Environment,Northeast Normal University,Changchun,China [3]Department of Integrative Biology and Physiology,University of California,Los Angeles,California,USA [4]Collage of Animal Science and Technology,Jilin Agricultural University,Changchun,China [5]Key Laboratory for Vegetation Ecology,Ministry of Education,Northeast Normal University,Changchun,China

出　　处：《Integrative Zoology》2022年第5期741-751,共11页整合动物学（英文版）

基　　金：supported by the National Natural Science Foundation of China(Grant Nos.31770429 and 31670390);the Natural Science Foundation of Jilin(Grant No.20180101263JC);the Program of Introducing Talents to Universities(Grant No.B16011);was also named the Project 111,and the National Program for“1000 Talent Plan for High-Level Foreign Experts”was from Organization Department of the CPC Central Committee(Grant No.WQ20142200259).

摘　　要：Acquiring clear acoustic signals is critical for the analysis of animal vocalizations.Bioacoustics studies commonly face the problem of overlapping signals,which can impede the structural identification of vocal units,but there is currently no satisfactory solution.This study presents a bi-directional long short-term memory network to separate overlapping echolocation-communication calls of 6 different bat species and reconstruct waveforms.The separation quality was evaluated using 7 temporal-spectrum parameters.All the echolocation pulses and syllables of communication calls in the overlapping signals were separated and parameter comparisons showed no significant difference and negligible deviation between the extracted and original calls.Clustering analysis was conducted with separated echolocation calls from each bat species to provide an example of practical application of the separated and reconstructed calls.The result of clustering analysis showed high corrected rand index(82.79%),suggesting the reconstructed waveforms could be reliably used for species classification.These results demonstrate a convenient and automated approach for separating overlapping calls.The study extends the application of deep neural networks to separate overlapping animal sounds.

关 键 词：bat vocalizations BIOACOUSTICS deep neural networks overlapping calls sound separation 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]