Holger Klinck
John W. Fitzpatrick Director, K. Lisa Yang Center for Conservation Bioacoustics
As director of the K. Lisa Yang Center for Conservation Bioacoustics, one of my goals is to enable researchers around the globe to acoustically monitor habitats and wildlife at large spatial scales. My current research focuses on the development and application of hardware and software tools to acoustically monitor terrestrial and marine ecosystems and biodiversity. I am also studying the impacts of anthropogenic noise on the vocal and locomotive behavior of animals. I advise several undergraduate and graduate students at Cornell and Oregon State University (OSU) and I regularly teach national and international bioacoustics classes.
Before moving to the United States for a postdoctoral position at OSU, I was a Ph.D. student at the Alfred Wegener Institute for Polar and Marine Research in Germany. My graduate work focused on the development of the Perennial Acoustic Observatory in the Antarctic Ocean and the study of the vocal behavior of leopard seals (coolest animals ever!) I am a full member of the Acoustical Society of America (ASA), responsible for the ASA technical committee Animal Bioacoustics website. I joined the Cornell Lab as the director of the Bioacoustics Research Program in 2016. I am also a Faculty Fellow with the Atkinson Center for a Sustainable Future at Cornell University. In addition, I hold an Adjunct Assistant Professor position at Oregon State University (OSU), where I lead the Research Collective for Applied Acoustics.
Beyond the Lab
I am an avid college and professional sports fan. My hobbies include running, sailing, and tinkering with gadgets. My wife Karolin and I live in Ithaca and enjoy hiking with our two Australian shepherd dogs, Lilly and Sammy.
Publications
2804417
Klinck
items
1
10
date
desc
title
2543
https://www.birds.cornell.edu/home/wp-content/plugins/zotpress/
%7B%22status%22%3A%22success%22%2C%22updateneeded%22%3Afalse%2C%22instance%22%3A%22zotpress-d90799182316d4140433335ae9599567%22%2C%22meta%22%3A%7B%22request_last%22%3A0%2C%22request_next%22%3A0%2C%22used_cache%22%3Atrue%7D%2C%22data%22%3A%5B%7B%22key%22%3A%22N54GPV9A%22%2C%22library%22%3A%7B%22id%22%3A2804417%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22McGinn%20et%20al.%22%2C%22parsedDate%22%3A%222023-05-01%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20padding-left%3A%201em%3B%20text-indent%3A-1em%3B%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%3EMcGinn%2C%20K.%2C%20S.%20Kahl%2C%20M.%20Z.%20Peery%2C%20H.%20Klinck%2C%20and%20C.%20M.%20Wood%20%282023%29.%20%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fwww.sciencedirect.com%5C%2Fscience%5C%2Farticle%5C%2Fpii%5C%2FS1574954123000249%27%3EFeature%20embeddings%20from%20the%20BirdNET%20algorithm%20provide%20insights%20into%20avian%20ecology%3C%5C%2Fa%3E.%20Ecological%20Informatics%2074%3A101995.%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Feature%20embeddings%20from%20the%20BirdNET%20algorithm%20provide%20insights%20into%20avian%20ecology%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Kate%22%2C%22lastName%22%3A%22McGinn%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Stefan%22%2C%22lastName%22%3A%22Kahl%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22M.%20Zachariah%22%2C%22lastName%22%3A%22Peery%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Holger%22%2C%22lastName%22%3A%22Klinck%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Connor%20M.%22%2C%22lastName%22%3A%22Wood%22%7D%5D%2C%22abstractNote%22%3A%22Bioacoustics%20has%20become%20widely%20used%20in%20the%20study%20of%20acoustically%20active%20animals%2C%20and%20machine%20learning%20algorithms%20have%20emerged%20as%20efficient%20and%20effective%20strategies%20to%20identify%20species%20vocalizations.%20Current%20applications%20of%20machine%20learning%20in%20bioacoustics%20often%20identify%20acoustic%20events%20to%20the%20species-level%20but%20fail%20to%20capture%20the%20complex%20acoustic%20repertoires%20animals%20use%20to%20communicate%2C%20which%20can%20inform%20habitat%20associations%2C%20demography%2C%20behavior%2C%20and%20the%20life%20history%20of%20cryptic%20species.%20The%20penultimate%20layer%20of%20most%20machine%20learning%20algorithms%20results%20in%20a%20vector%20of%20numbers%20describing%20the%20input%2C%20called%20feature%20embeddings.%20Here%2C%20we%20demonstrate%20that%20the%20feature%20embeddings%20generated%20by%20the%20BirdNET%20algorithm%20can%20enable%20within-species%20classifications%20of%20acoustic%20events.%20First%2C%20we%20successfully%20differentiated%20adult%20and%20juvenile%20Great%20Gray%20Owls%3B%20second%2C%20we%20identified%20three%20unique%20sounds%20associated%20with%20Great%20Spotted%20Woodpeckers%20%28series%20call%2C%20alarm%20call%2C%20and%20drumming%29.%20These%20applications%20of%20BirdNET%20feature%20embeddings%20suggest%20that%20researchers%20can%20classify%20vocalizations%20into%20groups%20when%20group%20membership%20is%20unknown%2C%20and%20that%20within-species%20grouping%20is%20possible%20even%20when%20target%20signals%20are%20extremely%20rare.%20These%20applications%20of%20a%20relatively%20%5Cu201cblack-box%5Cu201d%20aspect%20of%20machine%20learning%20algorithms%20can%20be%20used%20to%20derive%20ecologically%20informative%20acoustic%20classifications%2C%20which%20can%20inform%20the%20conservation%20of%20cryptic%20and%20otherwise%20difficult%20to%20study%20species.%22%2C%22date%22%3A%222023-05-01%22%2C%22language%22%3A%22en%22%2C%22DOI%22%3A%2210.1016%5C%2Fj.ecoinf.2023.101995%22%2C%22ISSN%22%3A%221574-9541%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.sciencedirect.com%5C%2Fscience%5C%2Farticle%5C%2Fpii%5C%2FS1574954123000249%22%2C%22collections%22%3A%5B%229BUGNQUJ%22%5D%2C%22dateModified%22%3A%222023-05-18T14%3A49%3A57Z%22%7D%7D%2C%7B%22key%22%3A%22A6ANAY54%22%2C%22library%22%3A%7B%22id%22%3A2804417%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Kennedy%20et%20al.%22%2C%22parsedDate%22%3A%222023-02-15%22%2C%22numChildren%22%3A2%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20padding-left%3A%201em%3B%20text-indent%3A-1em%3B%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%3EKennedy%2C%20A.%20G.%2C%20A.%20H.%20Ahmad%2C%20H.%20Klinck%2C%20L.%20M.%20Johnson%2C%20and%20D.%20J.%20Clink%20%282023%29.%20%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fonlinelibrary.wiley.com%5C%2Fdoi%5C%2Fabs%5C%2F10.1111%5C%2Fbtp.13205%27%3EEvidence%20for%20acoustic%20niche%20partitioning%20depends%20on%20the%20temporal%20scale%20in%20two%20sympatric%20Bornean%20hornbill%20species%3C%5C%2Fa%3E.%20Biotropica%20n%5C%2Fa.%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Evidence%20for%20acoustic%20niche%20partitioning%20depends%20on%20the%20temporal%20scale%20in%20two%20sympatric%20Bornean%20hornbill%20species%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Amy%20G.%22%2C%22lastName%22%3A%22Kennedy%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Abdul%20Hamid%22%2C%22lastName%22%3A%22Ahmad%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Holger%22%2C%22lastName%22%3A%22Klinck%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Lynn%20M.%22%2C%22lastName%22%3A%22Johnson%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Dena%20J.%22%2C%22lastName%22%3A%22Clink%22%7D%5D%2C%22abstractNote%22%3A%22Understanding%20niche%20partitioning%20of%20closely%20related%20sympatric%20species%20is%20a%20fundamental%20goal%20in%20ecology.%20Acoustic%20communication%20space%20can%20be%20considered%20a%20resource%2C%20and%20the%20acoustic%20niche%20hypothesis%20posits%20that%20competition%20between%20species%20may%20lead%20to%20partitioning%20of%20communication%20space.%20Here%2C%20we%20compare%20the%20calling%20behavior%20of%20two%20sympatric%20Bornean%20hornbill%20species%5Cu2014the%20rhinoceros%20hornbill%20%28Buceros%20rhinoceros%29%20and%20the%20helmeted%20hornbill%20%28Rhinoplax%20vigil%29%5Cu2014to%20test%20for%20evidence%20of%20acoustic%20niche%20partitioning.%20Both%20hornbill%20species%20emit%20calls%20heard%20over%20many%20kilometers%20and%20have%20similar%20habitat%20preferences%20which%20is%20predicted%20to%20result%20in%20interspecific%20competition.%20We%20collected%20acoustic%20data%20on%20sympatric%20populations%20of%20both%20hornbill%20species%20using%2010%20autonomous%20recording%20units%20in%20Danum%20Valley%20Conservation%20Area%2C%20Sabah%2C%20Malaysia.%20We%20found%20that%20there%20was%20substantial%20spectral%20overlap%20between%20the%20calls%20of%20the%20two%20species%2C%20indicating%20the%20potential%20for%20competition%20for%20acoustic%20space.%20To%20test%20for%20evidence%20of%20acoustic%20niche%20partitioning%2C%20we%20investigated%20spatial%20and%20temporal%20patterns%20of%20calling%20in%20each%20species.%20Both%20hornbills%20were%20strictly%20diurnal%20and%20called%20throughout%20the%20day%2C%20and%20we%20were%20equally%20likely%20to%20detect%20both%20species%20at%20each%20of%20our%20recorders.%20We%20did%20not%20find%20evidence%20of%20temporal%20acoustic%20avoidance%20at%20a%20relatively%20coarse%20timescale%20when%20we%20divided%20our%20dataset%20into%201%20h%20bins%2C%20but%20we%20did%20find%20evidence%20of%20temporal%20acoustic%20avoidance%20at%20a%20finer%20timescale%20when%20we%20used%20null%20models%20to%20compare%20the%20observed%20duration%20of%20overlapping%20calls%20to%20the%20expected%20amount%20of%20overlap%20due%20to%20chance.%20Abstract%20in%20Malay%20is%20available%20with%20online%20material.%22%2C%22date%22%3A%2215%20February%202023%22%2C%22language%22%3A%22en%22%2C%22DOI%22%3A%2210.1111%5C%2Fbtp.13205%22%2C%22ISSN%22%3A%221744-7429%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fonlinelibrary.wiley.com%5C%2Fdoi%5C%2Fabs%5C%2F10.1111%5C%2Fbtp.13205%22%2C%22collections%22%3A%5B%229BUGNQUJ%22%5D%2C%22dateModified%22%3A%222023-05-18T14%3A49%3A57Z%22%7D%7D%2C%7B%22key%22%3A%229UN88BI8%22%2C%22library%22%3A%7B%22id%22%3A2804417%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Barlow%20et%20al.%22%2C%22parsedDate%22%3A%222023%22%2C%22numChildren%22%3A3%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20padding-left%3A%201em%3B%20text-indent%3A-1em%3B%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%3EBarlow%2C%20D.%20R.%2C%20H.%20Klinck%2C%20D.%20Ponirakis%2C%20T.%20A.%20Branch%2C%20and%20L.%20G.%20Torres%20%282023%29.%20%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fonlinelibrary.wiley.com%5C%2Fdoi%5C%2Fabs%5C%2F10.1002%5C%2Fece3.9770%27%3EEnvironmental%20conditions%20and%20marine%20heatwaves%20influence%20blue%20whale%20foraging%20and%20reproductive%20effort%3C%5C%2Fa%3E.%20Ecology%20and%20Evolution%2013%3Ae9770.%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Environmental%20conditions%20and%20marine%20heatwaves%20influence%20blue%20whale%20foraging%20and%20reproductive%20effort%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Dawn%20R.%22%2C%22lastName%22%3A%22Barlow%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Holger%22%2C%22lastName%22%3A%22Klinck%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Dimitri%22%2C%22lastName%22%3A%22Ponirakis%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Trevor%20A.%22%2C%22lastName%22%3A%22Branch%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Leigh%20G.%22%2C%22lastName%22%3A%22Torres%22%7D%5D%2C%22abstractNote%22%3A%22Animal%20behavior%20is%20motivated%20by%20the%20fundamental%20need%20to%20feed%20and%20reproduce%2C%20and%20these%20behaviors%20can%20be%20inferred%20from%20spatiotemporal%20variations%20in%20biological%20signals%20such%20as%20vocalizations.%20Yet%2C%20linking%20foraging%20and%20reproductive%20effort%20to%20environmental%20drivers%20can%20be%20challenging%20for%20wide-ranging%20predator%20species.%20Blue%20whales%20are%20acoustically%20active%20marine%20predators%20that%20produce%20two%20distinct%20vocalizations%3A%20song%20and%20D%20calls.%20We%20examined%20environmental%20correlates%20of%20these%20vocalizations%20using%20continuous%20recordings%20from%20five%20hydrophones%20in%20the%20South%20Taranaki%20Bight%20region%20of%20Aotearoa%20New%20Zealand%20to%20investigate%20call%20behavior%20relative%20to%20ocean%20conditions%20and%20infer%20life%20history%20patterns.%20D%20calls%20were%20strongly%20correlated%20with%20oceanographic%20drivers%20of%20upwelling%20in%20spring%20and%20summer%2C%20indicating%20associations%20with%20foraging%20effort.%20In%20contrast%2C%20song%20displayed%20a%20highly%20seasonal%20pattern%20with%20peak%20intensity%20in%20fall%2C%20which%20aligned%20with%20the%20timing%20of%20conception%20inferred%20from%20whaling%20records.%20Finally%2C%20during%20a%20marine%20heatwave%2C%20reduced%20foraging%20%28inferred%20from%20D%20calls%29%20was%20followed%20by%20lower%20reproductive%20effort%20%28inferred%20from%20song%20intensity%29.%22%2C%22date%22%3A%222023%22%2C%22language%22%3A%22en%22%2C%22DOI%22%3A%2210.1002%5C%2Fece3.9770%22%2C%22ISSN%22%3A%222045-7758%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fonlinelibrary.wiley.com%5C%2Fdoi%5C%2Fabs%5C%2F10.1002%5C%2Fece3.9770%22%2C%22collections%22%3A%5B%229BUGNQUJ%22%5D%2C%22dateModified%22%3A%222023-05-18T14%3A49%3A57Z%22%7D%7D%2C%7B%22key%22%3A%228D7C4WWA%22%2C%22library%22%3A%7B%22id%22%3A2804417%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Clink%20et%20al.%22%2C%22parsedDate%22%3A%222023%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20padding-left%3A%201em%3B%20text-indent%3A-1em%3B%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%3EClink%2C%20D.%20J.%2C%20I.%20Kier%2C%20A.%20H.%20Ahmad%2C%20and%20H.%20Klinck%20%282023%29.%20%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fwww.frontiersin.org%5C%2Farticles%5C%2F10.3389%5C%2Ffevo.2023.1071640%27%3EA%20workflow%20for%20the%20automated%20detection%20and%20classification%20of%20female%20gibbon%20calls%20from%20long-term%20acoustic%20recordings%3C%5C%2Fa%3E.%20Frontiers%20in%20Ecology%20and%20Evolution%2011.%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22A%20workflow%20for%20the%20automated%20detection%20and%20classification%20of%20female%20gibbon%20calls%20from%20long-term%20acoustic%20recordings%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Dena%20J.%22%2C%22lastName%22%3A%22Clink%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Isabel%22%2C%22lastName%22%3A%22Kier%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Abdul%20Hamid%22%2C%22lastName%22%3A%22Ahmad%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Holger%22%2C%22lastName%22%3A%22Klinck%22%7D%5D%2C%22abstractNote%22%3A%22Passive%20acoustic%20monitoring%20%28PAM%29%20allows%20for%20the%20study%20of%20vocal%20animals%20on%20temporal%20and%20spatial%20scales%20difficult%20to%20achieve%20using%20only%20human%20observers.%20Recent%20improvements%20in%20recording%20technology%2C%20data%20storage%2C%20and%20battery%20capacity%20have%20led%20to%20increased%20use%20of%20PAM.%20One%20of%20the%20main%20obstacles%20in%20implementing%20wide-scale%20PAM%20programs%20is%20the%20lack%20of%20open-source%20programs%20that%20efficiently%20process%20terabytes%20of%20sound%20recordings%20and%20do%20not%20require%20large%20amounts%20of%20training%20data.%20Here%20we%20describe%20a%20workflow%20for%20detecting%2C%20classifying%2C%20and%20visualizing%20female%20Northern%20grey%20gibbon%20calls%20in%20Sabah%2C%20Malaysia.%20Our%20approach%20detects%20sound%20events%20using%20band-limited%20energy%20summation%20and%20does%20binary%20classification%20of%20these%20events%20%28gibbon%20female%20or%20not%29%20using%20machine%20learning%20algorithms%20%28support%20vector%20machine%20and%20random%20forest%29.%20We%20then%20applied%20an%20unsupervised%20approach%20%28affinity%20propagation%20clustering%29%20to%20see%20if%20we%20could%20further%20differentiate%20between%20true%20and%20false%20positives%20or%20the%20number%20of%20gibbon%20females%20in%20our%20dataset.%20We%20used%20this%20workflow%20to%20address%20three%20questions%3A%20%281%29%20does%20this%20automated%20approach%20provide%20reliable%20estimates%20of%20temporal%20patterns%20of%20gibbon%20calling%20activity%3B%20%282%29%20can%20unsupervised%20approaches%20be%20applied%20as%20a%20post-processing%20step%20to%20improve%20the%20performance%20of%20the%20system%3B%20and%20%283%29%20can%20unsupervised%20approaches%20be%20used%20to%20estimate%20how%20many%20female%20individuals%20%28or%20clusters%29%20there%20are%20in%20our%20study%20area%3F%20We%20found%20that%20performance%20plateaued%20with%20%3E160%20clips%20of%20training%20data%20for%20each%20of%20our%20two%20classes.%20Using%20optimized%20settings%2C%20our%20automated%20approach%20achieved%20a%20satisfactory%20performance%20%28F1%20score%5Cu2009~%5Cu200980%25%29.%20The%20unsupervised%20approach%20did%20not%20effectively%20differentiate%20between%20true%20and%20false%20positives%20or%20return%20clusters%20that%20appear%20to%20correspond%20to%20the%20number%20of%20females%20in%20our%20study%20area.%20Our%20results%20indicate%20that%20more%20work%20needs%20to%20be%20done%20before%20unsupervised%20approaches%20can%20be%20reliably%20used%20to%20estimate%20the%20number%20of%20individual%20animals%20occupying%20an%20area%20from%20PAM%20data.%20Future%20work%20applying%20these%20methods%20across%20sites%20and%20different%20gibbon%20species%20and%20comparisons%20to%20deep%20learning%20approaches%20will%20be%20crucial%20for%20future%20gibbon%20conservation%20initiatives%20across%20Southeast%20Asia.%22%2C%22date%22%3A%222023%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%222296-701X%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.frontiersin.org%5C%2Farticles%5C%2F10.3389%5C%2Ffevo.2023.1071640%22%2C%22collections%22%3A%5B%229BUGNQUJ%22%5D%2C%22dateModified%22%3A%222023-05-18T14%3A49%3A57Z%22%7D%7D%2C%7B%22key%22%3A%22KWC8Y8HA%22%2C%22library%22%3A%7B%22id%22%3A2804417%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Li%20et%20al.%22%2C%22parsedDate%22%3A%222023%22%2C%22numChildren%22%3A2%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20padding-left%3A%201em%3B%20text-indent%3A-1em%3B%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%3ELi%2C%20P.%2C%20M.%20A.%20Roch%2C%20H.%20Klinck%2C%20E.%20Fleishman%2C%20D.%20Gillespie%2C%20E.-M.%20Nosal%2C%20Y.%20Shiu%2C%20and%20X.%20Liu%20%282023%29.%20Learning%20Stage-wise%20GANs%20for%20Whistle%20Extraction%20in%20Time-Frequency%20Spectrograms.%20IEEE%20Transactions%20on%20Multimedia%3A1%26%23x2013%3B13.%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Learning%20Stage-wise%20GANs%20for%20Whistle%20Extraction%20in%20Time-Frequency%20Spectrograms%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Pu%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Marie%20A.%22%2C%22lastName%22%3A%22Roch%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Holger%22%2C%22lastName%22%3A%22Klinck%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Erica%22%2C%22lastName%22%3A%22Fleishman%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Douglas%22%2C%22lastName%22%3A%22Gillespie%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Eva-Marie%22%2C%22lastName%22%3A%22Nosal%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yu%22%2C%22lastName%22%3A%22Shiu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xiaobai%22%2C%22lastName%22%3A%22Liu%22%7D%5D%2C%22abstractNote%22%3A%22Whistle%20contour%20extraction%20aims%20to%20derive%20animal%20whistles%20from%20time-frequency%20spectrograms%20as%20polylines.%20For%20toothed%20whales%2C%20whistle%20extraction%20results%20can%20serve%20as%20the%20basis%20for%20analyzing%20animal%20abundance%2C%20species%20identity%2C%20and%20social%20activities.%20During%20the%20last%20few%20decades%2C%20as%20long-term%20recording%20systems%20have%20become%20affordable%2C%20automated%20whistle%20extraction%20algorithms%20were%20proposed%20to%20process%20large%20volumes%20of%20recording%20data.%20Recently%2C%20a%20deep%20learning-based%20method%20demonstrated%20superior%20performance%20in%20extracting%20whistles%20under%20varying%20noise%20conditions.%20However%2C%20training%20such%20networks%20requires%20a%20large%20amount%20of%20labor-intensive%20annotation%2C%20which%20is%20not%20available%20for%20many%20species.%20To%20overcome%20this%20limitation%2C%20we%20present%20a%20framework%20of%20stage-wise%20generative%20adversarial%20networks%20%28GANs%29%2C%20which%20compile%20new%20whistle%20data%20suitable%20for%20deep%20model%20training%20via%20three%20stages%3A%20generation%20of%20background%20noise%20in%20the%20spectrogram%2C%20generation%20of%20whistle%20contours%2C%20and%20generation%20of%20whistle%20signals.%20By%20separating%20the%20generation%20of%20different%20components%20in%20the%20samples%2C%20our%20framework%20composes%20visually%20promising%20whistle%20data%20and%20labels%20even%20when%20few%20expert%20annotated%20data%20are%20available.%20Regardless%20of%20the%20amount%20of%20human-annotated%20data%2C%20the%20proposed%20data%20augmentation%20framework%20leads%20to%20a%20consistent%20improvement%20in%20performance%20of%20the%20whistle%20extraction%20model%2C%20with%20a%20maximum%20increase%20of%201.69%20in%20the%20whistle%20extraction%20mean%20F1-score.%20Our%20stage-wise%20GAN%20also%20surpasses%20one%20single%20GAN%20in%20improving%20whistle%20extraction%20models%20with%20augmented%20data.%20The%20data%20and%20code%20will%20be%20available%20at%20https%3A%5C%2F%5C%2Fgithub.com%5C%2FPaul-LiPu%5C%2FCompositeGAN_WhistleAugment.%22%2C%22date%22%3A%222023%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%2210.1109%5C%2FTMM.2023.3251109%22%2C%22ISSN%22%3A%221941-0077%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%229BUGNQUJ%22%5D%2C%22dateModified%22%3A%222023-05-18T14%3A49%3A57Z%22%7D%7D%2C%7B%22key%22%3A%22E99EXTBY%22%2C%22library%22%3A%7B%22id%22%3A2804417%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Rameau%20et%20al.%22%2C%22parsedDate%22%3A%222022-12-19%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20padding-left%3A%201em%3B%20text-indent%3A-1em%3B%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%3ERameau%2C%20A.%2C%20K.%20Andreadis%2C%20V.%20Ganesan%2C%20M.%20S.%20Lachs%2C%20T.%20Rosen%2C%20F.%20Wang%2C%20A.%20Maddox%2C%20H.%20Klinck%2C%20S.%20M.%20Khosla%2C%20C.%20F.%20de%20Luzan%2C%20and%20S.%20Madhusudhana%20%282022%29.%20%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fonlinelibrary.wiley.com%5C%2Fdoi%5C%2Fabs%5C%2F10.1002%5C%2Flary.30525%27%3EAcoustic%20screening%20of%20the%20%5C%22wet%20voice%5C%22%3A%20proof%20of%20concept%20in%20an%20ex%20vivo%20canine%20laryngeal%20model%3C%5C%2Fa%3E.%20The%20Laryngoscope%20n%5C%2Fa.%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Acoustic%20screening%20of%20the%20%5C%22wet%20voice%5C%22%3A%20proof%20of%20concept%20in%20an%20ex%20vivo%20canine%20laryngeal%20model%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Ana%5Cu00efs%22%2C%22lastName%22%3A%22Rameau%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Katerina%22%2C%22lastName%22%3A%22Andreadis%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Vinayak%22%2C%22lastName%22%3A%22Ganesan%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Mark%20S%22%2C%22lastName%22%3A%22Lachs%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Tony%22%2C%22lastName%22%3A%22Rosen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Fei%22%2C%22lastName%22%3A%22Wang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Alexandra%22%2C%22lastName%22%3A%22Maddox%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Holger%22%2C%22lastName%22%3A%22Klinck%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Sid%20M.%22%2C%22lastName%22%3A%22Khosla%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Charles%20Farbos%22%2C%22lastName%22%3A%22de%20Luzan%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Shyam%22%2C%22lastName%22%3A%22Madhusudhana%22%7D%5D%2C%22abstractNote%22%3A%22Background%20Current%20protocols%20for%20bedside%20swallow%20evaluation%20have%20high%20rates%20of%20false%20negative%20results.%20Though%20experts%20are%20not%20consistently%20able%20to%20screen%20for%20aspiration%20risk%20by%20assessing%20vocal%20quality%2C%20there%20is%20emerging%20evidence%20that%20vocal%20acoustic%20parameters%20are%20significantly%20different%20in%20patients%20at%20risk%20of%20aspiration.%20Herein%2C%20we%20aimed%20to%20determine%20whether%20the%20presence%20of%20material%20on%20the%20vocal%20folds%20in%20an%20excised%20canine%20laryngeal%20model%20may%20have%20an%20impact%20on%20acoustic%20and%20aerodynamic%20measures.%20Methods%20Two%20ex%20vivo%20canine%20larynges%20were%20tested.%20Three%20liquids%20of%20different%20viscosities%20%281%3A100%20diluted%20glycerin%2C%20pure%20glycerin%2C%20and%20honey-thick%20Varibar%29%20were%20placed%20on%20the%20vocal%20folds%20at%20a%20constant%20volume.%20Acoustic%20and%20aerodynamic%20measures%20were%20obtained%20in%20both%20adducted%20and%20abducted%20vocal%20fold%20configurations.%20Intraglottal%20high-speed%20imaging%20was%20used%20to%20approximate%20the%20maximum%20divergence%20angle%20of%20the%20larynges%20in%20the%20studied%20conditions%20and%20examine%20its%20relationship%20to%20vocal%20efficiency%20%28VE%29%20and%20acoustic%20measures.%20Results%20In%20glottic%20insufficiency%20conditions%20only%2C%20we%20found%20that%20several%20acoustic%20parameters%20could%20predict%20the%20presence%20of%20material%20on%20the%20vocal%20folds.%20Based%20on%20the%20combination%20of%20the%20aerodynamic%20and%20acoustic%20data%2C%20we%20found%20that%20decreased%20spectral%20energy%20in%20the%20higher%20harmonics%20was%20associated%20with%20decreased%20VE%20in%20the%20presence%20of%20material%20on%20the%20vocal%20folds%20and%5C%2For%20glottic%20insufficiency.%20Conclusion%20Decreased%20spectral%20energy%20in%20the%20higher%20harmonics%20of%20the%20voice%20was%20found%20to%20be%20a%20potential%20biomarker%20of%20swallowing%20dysfunction%2C%20as%20it%20correlates%20with%20decreased%20vocal%20efficiency%20due%20to%20material%20on%20the%20vocal%20folds%20and%5C%2For%20glottic%20insufficiency%2C%20both%20of%20which%20are%20known%20risk%20factors%20for%20aspiration.%20Level%20of%20Evidence%20N%5C%2FA%20Laryngoscope%2C%202022%22%2C%22date%22%3A%2219%20December%202022%22%2C%22language%22%3A%22en%22%2C%22DOI%22%3A%2210.1002%5C%2Flary.30525%22%2C%22ISSN%22%3A%221531-4995%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fonlinelibrary.wiley.com%5C%2Fdoi%5C%2Fabs%5C%2F10.1002%5C%2Flary.30525%22%2C%22collections%22%3A%5B%22X6WG2SBA%22%2C%229BUGNQUJ%22%5D%2C%22dateModified%22%3A%222023-05-18T18%3A13%3A24Z%22%7D%7D%2C%7B%22key%22%3A%22D3IM6WYK%22%2C%22library%22%3A%7B%22id%22%3A2804417%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Ghani%20et%20al.%22%2C%22parsedDate%22%3A%222022-12-06%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20padding-left%3A%201em%3B%20text-indent%3A-1em%3B%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%3EGhani%2C%20B.%2C%20H.%20Klinck%2C%20and%20S.%20Hallerberg%20%282022%29.%20%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fwww.sciencedirect.com%5C%2Fscience%5C%2Farticle%5C%2Fpii%5C%2FS157495412200396X%27%3EClassification%20of%20group-specific%20variations%20in%20songs%20within%20House%20Wren%20species%20using%20machine%20learning%20models%3C%5C%2Fa%3E.%20Ecological%20Informatics%3A101946.%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Classification%20of%20group-specific%20variations%20in%20songs%20within%20House%20Wren%20species%20using%20machine%20learning%20models%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Burooj%22%2C%22lastName%22%3A%22Ghani%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Holger%22%2C%22lastName%22%3A%22Klinck%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Sarah%22%2C%22lastName%22%3A%22Hallerberg%22%7D%5D%2C%22abstractNote%22%3A%22Songbirds%20have%20shown%20variation%20in%20vocalizations%20across%20different%20populations%20and%20different%20geographical%20ranges.%20Such%20variations%20can%20over%20time%20lead%20to%20divergence%20in%20song%20characteristics%2C%20sometimes%20referred%20to%20as%20dialects.%20House%20Wren%20%28Troglodytes%20aedon%29%20is%20one%20such%20widely%20distributed%20bird%20species%20that%20has%20shown%20variation%20in%20its%20song%20characteristics%20within%20different%20populations.%20Traditionally%2C%20such%20studies%20have%20been%20conducted%20using%20manual%20approaches%20for%20classification.%20In%20this%20work%20we%20explore%20the%20use%20of%20machine%20learning%20models%20that%20can%20assist%20in%20performing%20classification%20of%20bird%20songs%20at%20a%20conspecific%20level.%20Two%20machine%20learning%20techniques%2C%20the%20random%20forest%20and%20a%20shallow%20feed%20forward%20neural%20network%2C%20are%20fed%20with%20pre-computed%20sound%20features%20to%20classify%20vocal%20variation%20in%20House%20Wren%20species%20across%20different%20reported%20population%20groups%20and%20latitudinal%20areas.%20A%20randomized%20approach%20is%20employed%20to%20create%20balanced%20subsets%20of%20sounds%20from%20different%20locations%20for%20repeated%20classification%20runs%20in%20order%20to%20provide%20a%20reliable%20estimate%20of%20performance.%20It%20is%20observed%20that%20such%20an%20automated%20approach%20is%20able%20to%20classify%20variations%20in%20songs%20within%20House%20Wren%20with%20high%20accuracy.%20We%20were%20also%20able%20to%20confirm%20the%20latitudinal%20variation%20of%20House%20Wren%20songs%20reported%20in%20previous%20studies.%20Given%20these%20results%2C%20we%20believe%2C%20such%20a%20purely%20data-driven%20way%20of%20analyzing%20bird%20songs%20in%20general%20can%20provide%20useful%20hints%20to%20biologists%20on%20where%20to%20look%20for%20interesting%20patterns%20in%20order%20to%20understand%20the%20evolutionary%20divergence%20in%20song%20characteristics.%22%2C%22date%22%3A%222022-12-06%22%2C%22language%22%3A%22en%22%2C%22DOI%22%3A%2210.1016%5C%2Fj.ecoinf.2022.101946%22%2C%22ISSN%22%3A%221574-9541%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.sciencedirect.com%5C%2Fscience%5C%2Farticle%5C%2Fpii%5C%2FS157495412200396X%22%2C%22collections%22%3A%5B%229BUGNQUJ%22%5D%2C%22dateModified%22%3A%222023-05-18T14%3A49%3A57Z%22%7D%7D%2C%7B%22key%22%3A%22W4QSHUIT%22%2C%22library%22%3A%7B%22id%22%3A2804417%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Conant%20et%20al.%22%2C%22parsedDate%22%3A%222022-12%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20padding-left%3A%201em%3B%20text-indent%3A-1em%3B%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%3EConant%2C%20P.%20C.%2C%20P.%20Li%2C%20X.%20Liu%2C%20H.%20Klinck%2C%20E.%20Fleishman%2C%20D.%20Gillespie%2C%20E.-M.%20Nosal%2C%20and%20M.%20A.%20Roch%20%282022%29.%20%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fasa.scitation.org%5C%2Fdoi%5C%2F10.1121%5C%2F10.0016631%27%3ESilbido%20profundo%3A%20An%20open%20source%20package%20for%20the%20use%20of%20deep%20learning%20to%20detect%20odontocete%20whistles%3C%5C%2Fa%3E.%20The%20Journal%20of%20the%20Acoustical%20Society%20of%20America%20152%3A3800%5Cu20133808.%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Silbido%20profundo%3A%20An%20open%20source%20package%20for%20the%20use%20of%20deep%20learning%20to%20detect%20odontocete%20whistles%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Peter%20C.%22%2C%22lastName%22%3A%22Conant%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Pu%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xiaobai%22%2C%22lastName%22%3A%22Liu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Holger%22%2C%22lastName%22%3A%22Klinck%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Erica%22%2C%22lastName%22%3A%22Fleishman%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Douglas%22%2C%22lastName%22%3A%22Gillespie%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Eva-Marie%22%2C%22lastName%22%3A%22Nosal%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Marie%20A.%22%2C%22lastName%22%3A%22Roch%22%7D%5D%2C%22abstractNote%22%3A%22This%20work%20presents%20an%20open-source%20matlab%20software%20package%20for%20exploiting%20recent%20advances%20in%20extracting%20tonal%20signals%20from%20large%20acoustic%20data%20sets.%20A%20whistle%20extraction%20algorithm%20published%20by%20Li%2C%20Liu%2C%20Palmer%2C%20Fleishman%2C%20Gillespie%2C%20Nosal%2C%20Shiu%2C%20Klinck%2C%20Cholewiak%2C%20Helble%2C%20and%20Roch%20%5B%282020%29.%20Proceedings%20of%20the%20International%20Joint%20Conference%20on%20Neural%20Networks%2C%20July%2019%5Cu201324%2C%20Glasgow%2C%20Scotland%2C%20p.%2010%5D%20is%20incorporated%20into%20silbido%2C%20an%20established%20software%20package%20for%20extraction%20of%20cetacean%20tonal%20calls.%20The%20precision%20and%20recall%20of%20the%20new%20system%20were%20over%2096%25%20and%20nearly%2080%25%2C%20respectively%2C%20when%20applied%20to%20a%20whistle%20extraction%20task%20on%20a%20challenging%20two-species%20subset%20of%20a%20conference-benchmark%20data%20set.%20A%20second%20data%20set%20was%20examined%20to%20assess%20whether%20the%20algorithm%20generalized%20to%20data%20that%20were%20collected%20across%20different%20recording%20devices%20and%20locations.%20These%20data%20included%20487%5Cu2009h%20of%20weakly%20labeled%2C%20towed%20array%20data%20collected%20in%20the%20Pacific%20Ocean%20on%20two%20National%20Oceanographic%20and%20Atmospheric%20Administration%20%28NOAA%29%20cruises.%20Labels%20for%20these%20data%20consisted%20of%20regions%20of%20toothed%20whale%20presence%20for%20at%20least%2015%20species%20that%20were%20based%20on%20visual%20and%20acoustic%20observations%20and%20not%20limited%20to%20whistles.%20Although%20the%20lack%20of%20per%20whistle-level%20annotations%20prevented%20measurement%20of%20precision%20and%20recall%2C%20there%20was%20strong%20concurrence%20of%20automatic%20detections%20and%20the%20NOAA%20annotations%2C%20suggesting%20that%20the%20algorithm%20generalizes%20well%20to%20new%20data.%22%2C%22date%22%3A%222022-12%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%2210.1121%5C%2F10.0016631%22%2C%22ISSN%22%3A%220001-4966%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fasa.scitation.org%5C%2Fdoi%5C%2F10.1121%5C%2F10.0016631%22%2C%22collections%22%3A%5B%229BUGNQUJ%22%5D%2C%22dateModified%22%3A%222023-05-18T14%3A49%3A57Z%22%7D%7D%2C%7B%22key%22%3A%2284DFDXU6%22%2C%22library%22%3A%7B%22id%22%3A2804417%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Fregosi%20et%20al.%22%2C%22parsedDate%22%3A%222022-10%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20padding-left%3A%201em%3B%20text-indent%3A-1em%3B%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%3EFregosi%2C%20S.%2C%20D.%20V.%20Harris%2C%20H.%20Matsumoto%2C%20D.%20K.%20Mellinger%2C%20S.%20W.%20Martin%2C%20B.%20Matsuyama%2C%20J.%20Barlow%2C%20and%20H.%20Klinck%20%282022%29.%20%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fasa.scitation.org%5C%2Fdoi%5C%2F10.1121%5C%2F10.0014793%27%3EDetection%20probability%20and%20density%20estimation%20of%20fin%20whales%20by%20a%20Seaglider%3C%5C%2Fa%3E.%20The%20Journal%20of%20the%20Acoustical%20Society%20of%20America%20152%3A2277%5Cu20132291.%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Detection%20probability%20and%20density%20estimation%20of%20fin%20whales%20by%20a%20Seaglider%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Selene%22%2C%22lastName%22%3A%22Fregosi%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Danielle%20V.%22%2C%22lastName%22%3A%22Harris%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Haruyoshi%22%2C%22lastName%22%3A%22Matsumoto%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22David%20K.%22%2C%22lastName%22%3A%22Mellinger%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Stephen%20W.%22%2C%22lastName%22%3A%22Martin%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Brian%22%2C%22lastName%22%3A%22Matsuyama%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jay%22%2C%22lastName%22%3A%22Barlow%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Holger%22%2C%22lastName%22%3A%22Klinck%22%7D%5D%2C%22abstractNote%22%3A%22A%20single-hydrophone%20ocean%20glider%20was%20deployed%20within%20a%20cabled%20hydrophone%20array%20to%20demonstrate%20a%20framework%20for%20estimating%20population%20density%20of%20fin%20whales%20%28Balaenoptera%20physalus%29%20from%20a%20passive%20acoustic%20glider.%20The%20array%20was%20used%20to%20estimate%20tracks%20of%20acoustically%20active%20whales.%20These%20tracks%20became%20detection%20trials%20to%20model%20the%20detection%20function%20for%20glider-recorded%20360-s%20windows%20containing%20fin%20whale%2020-Hz%20pulses%20using%20a%20generalized%20additive%20model.%20Detection%20probability%20was%20dependent%20on%20both%20horizontal%20distance%20and%20low-frequency%20glider%20flow%20noise.%20At%20the%20median%2040-Hz%20spectral%20level%20of%2097%5Cu2009dB%20re%201%20%5Cu03bcPa2%5C%2FHz%2C%20detection%20probability%20was%20near%20one%20at%20horizontal%20distance%20zero%20with%20an%20effective%20detection%20radius%20of%2017.1%5Cu2009km%20%5Bcoefficient%20of%20variation%20%28CV%29%5Cu2009%3D%5Cu20090.13%5D.%20Using%20estimates%20of%20acoustic%20availability%20and%20acoustically%20active%20group%20size%20from%20tagged%20and%20tracked%20fin%20whales%2C%20respectively%2C%20density%20of%20fin%20whales%20was%20estimated%20as%201.8%20whales%20per%201000%5Cu2009km2%20%28CV%5Cu2009%3D%5Cu20090.55%29.%20A%20plot%20sampling%20density%20estimate%20for%20the%20same%20area%20and%20time%2C%20estimated%20from%20array%20data%20alone%2C%20was%201.3%20whales%20per%201000%5Cu2009km2%20%28CV%5Cu2009%3D%5Cu20090.51%29.%20While%20the%20presented%20density%20estimates%20are%20from%20a%20small%20demonstration%20experiment%20and%20should%20be%20used%20with%20caution%2C%20the%20framework%20presented%20here%20advances%20our%20understanding%20of%20the%20potential%20use%20of%20gliders%20for%20cetacean%20density%20estimation.%22%2C%22date%22%3A%222022-10%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%2210.1121%5C%2F10.0014793%22%2C%22ISSN%22%3A%220001-4966%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fasa.scitation.org%5C%2Fdoi%5C%2F10.1121%5C%2F10.0014793%22%2C%22collections%22%3A%5B%22X6WG2SBA%22%5D%2C%22dateModified%22%3A%222022-10-27T17%3A02%3A38Z%22%7D%7D%2C%7B%22key%22%3A%22EMNRPEPE%22%2C%22library%22%3A%7B%22id%22%3A2804417%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Kahl%20et%20al.%22%2C%22parsedDate%22%3A%222022-09-05%22%2C%22numChildren%22%3A0%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20padding-left%3A%201em%3B%20text-indent%3A-1em%3B%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%3EKahl%2C%20S.%2C%20A.%20Navine%2C%20T.%20Denton%2C%20H.%20Klinck%2C%20P.%20Hart%2C%20H.%20Glotin%2C%20H.%20Go%5Cu00ebau%2C%20W.-P.%20Vellinga%2C%20R.%20Planqu%5Cu00e9%2C%20and%20A.%20Joly%20%282022%29.%20%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fhal.inrae.fr%5C%2Fhal-03791428%27%3EOverview%20of%20BirdCLEF%202022%3A%20Endangered%20bird%20species%20recognition%20in%20soundscape%20recordings%3C%5C%2Fa%3E.%20p.%201929.%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Overview%20of%20BirdCLEF%202022%3A%20Endangered%20bird%20species%20recognition%20in%20soundscape%20recordings%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Stefan%22%2C%22lastName%22%3A%22Kahl%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Amanda%22%2C%22lastName%22%3A%22Navine%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Tom%22%2C%22lastName%22%3A%22Denton%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Holger%22%2C%22lastName%22%3A%22Klinck%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Patrick%22%2C%22lastName%22%3A%22Hart%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Herv%5Cu00e9%22%2C%22lastName%22%3A%22Glotin%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Herv%5Cu00e9%22%2C%22lastName%22%3A%22Go%5Cu00ebau%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Willem-Pier%22%2C%22lastName%22%3A%22Vellinga%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Robert%22%2C%22lastName%22%3A%22Planqu%5Cu00e9%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Alexis%22%2C%22lastName%22%3A%22Joly%22%7D%5D%2C%22abstractNote%22%3A%22As%20the%20%5C%22extinction%20capital%20of%20the%20world%5C%22%2C%20Hawai%27i%20has%20lost%2068%25%20of%20its%20native%20bird%20species%2C%20the%20consequences%20of%20which%20can%20harm%20entire%20ecosystems.%20With%20physical%20monitoring%20difficult%2C%20scientists%20have%20turned%20to%20sound%20recordings%2C%20as%20this%20approach%20could%20provide%20a%20passive%2C%20low%20labor%2C%20and%20cost-effective%20strategy%20for%20monitoring%20endangered%20bird%20populations.%20Current%20methods%20for%20processing%20large%20bioacoustic%20datasets%20involve%20manual%20review%20of%20each%20recording.%20This%20requires%20specialized%20training%20and%20prohibitively%20large%20amounts%20of%20time.%20Recent%20advances%20in%20machine%20learning%20have%20made%20it%20possible%20to%20automatically%20identify%20bird%20songs%20for%20common%20species%20with%20ample%20training%20data.%20However%2C%20it%20remains%20challenging%20to%20develop%20such%20tools%20for%20rare%20and%20endangered%20species.%20The%20main%20goal%20of%20the%202022%20edition%20of%20BirdCLEF%20was%20to%20advance%20automated%20detection%20of%20rare%20and%20endangered%20bird%20species%20that%20lack%20large%20amounts%20of%20training%20data.%20The%20competition%20challenged%20participants%20to%20develop%20reliable%20analysis%20frameworks%20to%20detect%20and%20identify%20the%20vocalizations%20of%20rare%20bird%20species%20in%20continuous%20Hawaiian%20soundscapes%20utilizing%20limited%20training%20data.%22%2C%22date%22%3A%222022%5C%2F09%5C%2F05%22%2C%22proceedingsTitle%22%3A%22%22%2C%22conferenceName%22%3A%22CLEF%202022%20-%20Conference%20and%20Labs%20of%20the%20Evaluation%20Forum%22%2C%22language%22%3A%22en%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fhal.inrae.fr%5C%2Fhal-03791428%22%2C%22collections%22%3A%5B%229BUGNQUJ%22%5D%2C%22dateModified%22%3A%222023-05-18T14%3A49%3A57Z%22%7D%7D%5D%7D
McGinn, K., S. Kahl, M. Z. Peery, H. Klinck, and C. M. Wood (2023).
Feature embeddings from the BirdNET algorithm provide insights into avian ecology. Ecological Informatics 74:101995.
Barlow, D. R., H. Klinck, D. Ponirakis, T. A. Branch, and L. G. Torres (2023).
Environmental conditions and marine heatwaves influence blue whale foraging and reproductive effort. Ecology and Evolution 13:e9770.
Li, P., M. A. Roch, H. Klinck, E. Fleishman, D. Gillespie, E.-M. Nosal, Y. Shiu, and X. Liu (2023). Learning Stage-wise GANs for Whistle Extraction in Time-Frequency Spectrograms. IEEE Transactions on Multimedia:1–13.
Rameau, A., K. Andreadis, V. Ganesan, M. S. Lachs, T. Rosen, F. Wang, A. Maddox, H. Klinck, S. M. Khosla, C. F. de Luzan, and S. Madhusudhana (2022).
Acoustic screening of the "wet voice": proof of concept in an ex vivo canine laryngeal model. The Laryngoscope n/a.
Conant, P. C., P. Li, X. Liu, H. Klinck, E. Fleishman, D. Gillespie, E.-M. Nosal, and M. A. Roch (2022).
Silbido profundo: An open source package for the use of deep learning to detect odontocete whistles. The Journal of the Acoustical Society of America 152:3800–3808.
Fregosi, S., D. V. Harris, H. Matsumoto, D. K. Mellinger, S. W. Martin, B. Matsuyama, J. Barlow, and H. Klinck (2022).
Detection probability and density estimation of fin whales by a Seaglider. The Journal of the Acoustical Society of America 152:2277–2291.
Kahl, S., A. Navine, T. Denton, H. Klinck, P. Hart, H. Glotin, H. Goëau, W.-P. Vellinga, R. Planqué, and A. Joly (2022).
Overview of BirdCLEF 2022: Endangered bird species recognition in soundscape recordings. p. 1929.