Publications (5) View all
-
Article: Integrating computer-assistance and human-review to build richly annotated sound libraries.
[show abstract] [hide abstract]
ABSTRACT: A current challenge to the effective application of bioacoustic survey methods is the curation and sharing of large collections of sound and metadata, and their integration into a data-analysis workflow that includes computer-assistance and human-review. Computer-assistance affords the annotation of ever larger amounts of data and human-review provides for sufficient quality of annotation and creates a feedback mechanism that supports continuous learning through an ever-expanding collection of training data. I will present some solutions to the above challenges that are being implemented as part of the Bioacoustic Resource Network (BARN) project. BARN provides a web-based interface to annotated sound collections with a back-end computational engine based on XBAT. The solutions provided by the BARN platform contain both technical and social elements. In the technical arena, we consider problems of data-modeling for extensible annotation; strategies for the unique identification of data, derived annotations, and computational resources; the definition of workflow-oriented programming interfaces; and web-service approaches to the larger-scale usability of these annotation libraries. These technical approaches are complemented by network-supported social strategies such as public repositories, open-source licensing, and a developer network. Integration of these technical and social components promises to overcome current challenges and realize the potential of bioacoustic methods.The Journal of the Acoustical Society of America 10/2010; 128(4):2437. · 1.55 Impact Factor -
Article: Software for bioacoustic analysis of passive acoustic data.
[show abstract] [hide abstract]
ABSTRACT: Software analysis systems comprise an important stage in passive acoustic research. Here we compare and evaluate three animal acoustic analysis software systems: Ishmael, PAMGUARD, and XBAT. These packages are compared and evaluated for their capabilities at some of the common tasks in animal sound analysis: recording, display, detection, classification, measurement, localization, and tracking of animal sounds. They are also compared for their extensibility (how easy is it, say, to add a new detection algorithm), their ease of use (how hard is the software to learn to use, how quick to use it once one knows how), their hardware interfacing (what types of sound acquisition hardware can they receive sound from), their sound file interfaces (can they read everyone's sound files), their documentation, and their software interfaces (how well do they share data with other programs, other methodologies, like visual surveys). Special features of each system will also be discussed.The Journal of the Acoustical Society of America 05/2009; 125(4):2547. · 1.55 Impact Factor -
Article: Using image processing to detect and classify narrow-band cricket and frog calls.
[show abstract] [hide abstract]
ABSTRACT: An automatic call recognition (ACR) process is described that uses image processing techniques on spectrogram images to detect and classify constant-frequency cricket and frog calls recorded amidst a background of evening sounds found in a lowland Costa Rican rainforest. This process involves using image blur filters along with thresholding filters to isolate likely calling events. Features of these events, notably the event's central frequency, duration and bandwidth, along with the type of blur filter applied, are used with a Bayesian classifier to make identifications of the different calls. Of the 22 distinct sonotypes (calls presumed to be species-specific) recorded in the study site, 17 of them were recorded in high enough numbers to both train and test the classifier. The classifier approaches 100% true-positive accuracy for these 17 sonotypes, but also has a high false-negative rate (over 50% for 4 sonotypes). The very high true-positive accuracy of this process enables its use for monitoring singing crickets (and some frog species) in tropical forests.The Journal of the Acoustical Society of America 12/2006; 120(5 Pt 1):2950-7. · 1.55 Impact Factor -
Article: Notes and double knocks from Arkansas.
Science 10/2005; 309(5740):1489. · 31.20 Impact Factor -
Article: Nearest-neighbor techniques for automated monitoring of nocturnal flight calls.
[show abstract] [hide abstract]
ABSTRACT: Flight-calls are short vocalizations used primarily during nocturnal flight. Their observation provides a means for studying the timing, location, and composition of nocturnal migrations. As part of a three-year study the Cornell Lab of Ornithology is using autonomous recorders to sample flight-calls of nocturnal migrants in the Northeastern US. The resulting tens of thousands of hours of recording, make software-assisted detection and classification essential. Automatic processing and human evaluation have yielded a considerable collection of flight-calls, 5-1000 examples for 100 species. The many-class classification problem, along with the availability of many examples from most of the classes, and established (condensation and editing) and recent (metric-trees) techniques used in prototype-based classification nearest-neighbor techniques, have led us to develop nearest-neighbor based techniques and software to assist in the analysis of this data. We will present classification results on two examples, a set of thrushes (genera Catharus and Hylocichla, family Turdidae) consisting of six species and wood-warblers (family Parulidae) consisting of 48 species. The thrush flight-calls are visually and aurally distinctive, usually 100-400 ms in duration and occupy and the 2-5 kHz band. Wood-warbler flight-calls are typically between 20-100 ms in duration and occupy the 5-10 kHz, and are difficult for many experienced observers to distinguish.The Journal of the Acoustical Society of America 06/2008; 123(5):3101. · 1.55 Impact Factor
