Abstract
KLFromRecordingDays allows measurement of Kullback–Leibler (KL) distances between 2D probability distributions of vocal acoustic features. Greater KL distance measures reflect increased phonological divergence across the vocalizations compared. The software has been used to compare *.wav file recordings made by Sound Analysis Recorder 2011 of songbird vocalizations pre- and post-drug and surgical manipulations. Recordings from individual animals in *.wav format are first organized into subdirectories by recording day and then segmented into individual syllables uttered and acoustic features of these syllables using Sound Analysis Pro 2011 (SAP). KLFromRecordingDays uses syllable acoustic feature data output by SAP to a MySQL table to generate and compare “template” (typically pre-treatment) and “target” (typically post-treatment) probability distributions. These distributions are a series of virtual 2D plots of the duration of each syllable (as x-axis) to each of 13 other acoustic features measured by SAP for that syllable (as y-axes). Differences between “template” and “target” probability distributions for each acoustic feature are determined by calculating KL distance, a measure of divergence of the target 2D distribution pattern from that of the template. KL distances and the mean KL distance across all acoustic features are calculated for each recording day and output to an Excel spreadsheet. Resulting data for individual subjects may then be pooled across treatment groups and graphically summarized and used for statistical comparisons. Because SAP-generated MySQL files are accessed directly, data limits associated with spreadsheet output are avoided, and the totality of vocal output over weeks may be objectively analyzed all at once. The software has been useful for measuring drug effects on songbird vocalizations and assessing recovery from damage to regions of vocal motor cortex. It may be useful in studies employing other species, and as part of speech therapies tracking progress in producing distinct speech sounds in isolation.
Highlights
Speech disorders afflict approximately 4% of children by the time they reach six years of age, and are often secondary to other more severe developmental disorders including intractable childhood epilepsy and autism [1]
The goal in developing the current software was to integrate these approaches in a manner allowing efficient analysis of the totality of vocalizations produced over experiments lasting several weeks
The software described solves the problem of applying KL distance methods of quantifying phonology to the literally millions of acoustic feature measures calculated by Sound Analysis Pro (SAP) that are derived from every syllable recorded from animals over a multi-week experimental period
Summary
Speech disorders afflict approximately 4% of children by the time they reach six years of age, and are often secondary to other more severe developmental disorders including intractable childhood epilepsy and autism [1]. Very few animals learn a form of vocal communication in a manner similar to humans. Songbirds are among this small group of vocal learners. We have begun to develop a songbird, the zebra finch, as a laboratory animal model to evaluate drugs for effects to improve vocal learning and recovery from CNS damage. This model depends upon development of methods to objectively compare quality of vocalizations pre- and post-treatment over the course of weeks. The goal in developing the current software was to integrate these approaches in a manner allowing efficient analysis of the totality of vocalizations produced over experiments lasting several weeks
Sign-in/Register to view highlights & summary 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 callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
