Automatic acoustic identification of individuals in multiple species: improving identification across recording conditions

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68081766%3A_____%2F19%3A00504422" target="_blank" >RIV/68081766:_____/19:00504422 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/00216208:11310/19:10394305 RIV/60460709:41330/19:79543

Výsledek na webu

<a href="http://dx.doi.org/10.1098/rsif.2018.0940" target="_blank" >http://dx.doi.org/10.1098/rsif.2018.0940</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1098/rsif.2018.0940" target="_blank" >10.1098/rsif.2018.0940</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Automatic acoustic identification of individuals in multiple species: improving identification across recording conditions

Popis výsledku v původním jazyce

Many animals emit vocal sounds which, independently from the sounds' function, contain some individually distinctive signature. Thus the automatic recognition of individuals by sound is a potentially powerful tool for zoology and ecology research and practical monitoring. Here, we present a general automatic identification method that can work across multiple animal species with various levels of complexity in their communication systems. We further introduce new analysis techniques based on dataset manipulations that can evaluate the robustness and generality of a classifier. By using these techniques, we confirmed the presence of experimental confounds in situations resembling those from past studies. We introduce data manipulations that can reduce the impact of these confounds, compatible with any classifier. We suggest that assessment of confounds should become a standard part of future studies to ensure they do not report over-optimistic results. We provide annotated recordings used for analyses along with this study and we call for dataset sharing to be a common practice to enhance the development of methods and comparisons of results.

Název v anglickém jazyce

Automatic acoustic identification of individuals in multiple species: improving identification across recording conditions

Popis výsledku anglicky

Many animals emit vocal sounds which, independently from the sounds' function, contain some individually distinctive signature. Thus the automatic recognition of individuals by sound is a potentially powerful tool for zoology and ecology research and practical monitoring. Here, we present a general automatic identification method that can work across multiple animal species with various levels of complexity in their communication systems. We further introduce new analysis techniques based on dataset manipulations that can evaluate the robustness and generality of a classifier. By using these techniques, we confirmed the presence of experimental confounds in situations resembling those from past studies. We introduce data manipulations that can reduce the impact of these confounds, compatible with any classifier. We suggest that assessment of confounds should become a standard part of future studies to ensure they do not report over-optimistic results. We provide annotated recordings used for analyses along with this study and we call for dataset sharing to be a common practice to enhance the development of methods and comparisons of results.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10613 - Zoology

Projekt

<a href="/cs/project/GPP505%2F11%2FP572" target="_blank" >GPP505/11/P572: Informační hodnota specifických struktur ve zpěvu lindušky lesní: experimentální test hypotéz o funkci ptačího zpěvu</a><br>

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2019

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Journal of the Royal Society Interface

ISSN

1742-5689

e-ISSN

—

Svazek periodika

16

Číslo periodika v rámci svazku

153

Stát vydavatele periodika

GB - Spojené království Velké Británie a Severního Irska

Počet stran výsledku

13

Strana od-do

20180940

Kód UT WoS článku

000466431900007

EID výsledku v databázi Scopus

2-s2.0-85064721811