Automatic bird sound detection in long real-field recordings: Applications and tools Ilyas Potamitis a,⇑ , Stavros Ntalampiras b , Olaf Jahn c , Klaus Riede c a Technological Educational Institute of Crete, Department of Music Technology and Acoustics, Crete, Greece 1 b Department of Electronics & Information, Polytechnic of Milano, Italy c Zoological Research Museum Alexander Koenig, 53113 Bonn, Germany article info Article history: Received 8 January 2013 Received in revised form 13 August 2013 Accepted 8 January 2014 Keywords: Birdsong detection Bird recognition Computational ecology abstract The primary purpose for pursuing this research is to present a modular approach that enables reliable automatic bird species identification on the basis of their sound emissions in the field. A practical and complete computer-based framework is proposed to detect and time-stamp particular bird species in continuous real field recordings. Acoustic detection of avian sounds can be used for the automatized monitoring of multiple bird taxa and querying in long-term recordings for species of interest for research- ers, conservation practitioners, and decision makers, such as environmental indicator taxa and threa- tened species. This work describes two novel procedures and offers an open modular framework that detects and time-stamps online calls and songs of target bird species and is fast enough to report results in reasonable time for non-processed field recordings of many thousands files and is generic enough to accommodate any species. The framework is evaluated on two large corpora of real field data, targeting the calls and songs of American Robin Turdus migratorius, a Northamerican oscine passerine (true song- bird) and the Common Kingfisher Alcedo atthis, a non-passerine species with a wide distribution through- out Eurasia and North Africa. With the aim of promoting the widespread use of digital autonomous recording units (ARUs) and species recognition technologies the processing code and a large corpus of audio recordings is provided in order to enable other researchers to perform and assess comparative experiments. Ó 2014 Elsevier Ltd. All rights reserved. 1. Introduction In our rapidly changing world the monitoring of animal com- munities is becoming increasingly important. Reliable estimates of the range, population size, and population trends are critical for assessing the conservation status of the species. Only if we know their real population status, species-specific conservation measures can be implemented and extinctions can be avoided. However, the high costs of classical observer-based survey tech- niques and their temporal limitation are often a major problem for the protection of wildlife. A potential solution is the automated acoustic monitoring of sound-emitting animals, as they can pro- vide continuous real-time information on the presence/absence of target species and on the general status of the biodiversity of an area. Consequently, in recent years biologist started to use autono- mous recording units (ARUs) to survey different taxonomic groups of sound-producing animals, such as mammals [1], birds [2,3], amphibians [4], and insects [5]. Considering that these ARUs can be operated in 24/7 modus and that several recorders can be used simultaneously, huge amounts of audio data can be gathered in rel- atively short periods of time, meaning that it is usually not feasible for human experts to hear or visually inspect the complete sample of recordings. Thus (semi-)automatic processing of the sound files is a prerequisite for analyzing the information in a timely manner. The operation of autonomous remote audio recording stations and the automatic analysis of their data can assist decision making in a wide spectrum of areas, such as: (1) Monitoring of range shifts of animal species due to climate change. Greenhouse warming is projected to profoundly change the distribution pattern of plants and animals world- wide. For instance, the average distributional range of Euro- pean birds might shift nearly 550 km north-east by the end of this century [6–9]. In the same period, about 75% of the avian species might suffer range declines and the overlap of the current and future distribution might be only 40%. 0003-682X/$ - see front matter Ó 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.apacoust.2014.01.001 ⇑ Corresponding author. Address: Department of Music Technology & Acoustics, Technological Educational Institute of Crete, E. Daskalaki, Perivolia, Rethymno 74100, Crete, Greece. Tel.: +30 28310 21900. E-mail address: potamitis@staff.teicrete.gr (I. Potamitis). URL: http://scholar.google.com/citations?user=gWZ4dTUAAAAJ&hl=en I. Potamitis). 1 http://www.teicrete.gr/mta/en/index.php?q=node/32. Applied Acoustics 80 (2014) 1–9 Contents lists available at ScienceDirect Applied Acoustics journal homepage: www.elsevier.com/locate/apacoust