Abstract
Reno and colleagues (2013) recently published a first of itskind article describing their development and use of a novelalgorithm for the automatic detection of ultrasonic vocalizations(USVs). The WAAVES (Wav-file Automated Analysis of Vocaliza-tions Environment Specific) procedure is designed to effectivelydetect fixed-frequency (FF) and frequency modulated (FM) ultra-sonic vocalizations using MATLAB’s Signal and Image ProcessingToolbox.The authors should be applauded for their efforts. Indeed, mod-els of addiction, mood disorders, or pain and discomfort (amongstothers) can benefit from the automation of ultrasonic vocalizationdetection andscoring.However,effortstoautomateUSVdetectionhave beenrelativelyabsentfromtheliterature.Beyondthepreclin-ical value of USV detection, it seems that WAAVES is particularlyadvantageous for a number of other reasons: (1) MATLAB providesa quasiopen-sourcemethodforfurtherdevelopingdetectionalgo-rithms. Thus, the USV community as a whole might contribute inorder to improve the results of the detector should the authorschoose to publish the source code alongside the article. (2) Thedetector has been explicitly tuned for USVs in both the 22- and 50-kHz rangesandhasbeendesignedtofilterartifactscommontoUSVrecordings in drug abuse paradigms—artifacts that make a numberof commonbioacoustictechniqueslikeamplitudedetectorsnearlyuseless for drug abuse data. (3) The signal detection proceduresused in WAAVES might prove beneficial in other bioacoustic signaldetection applications (i.e. birdsong analysis) should the correctmodifications be made.Clearly, this novel procedure should be further evaluated andempirically tested by other laboratories. Indeed, the details pro-vided by the authors are somewhat sparse and the authors do not
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call