September, 2014

BioSpeech Inc. receives $617,118 for new NIH Phase I on automatic classification of rodent vocalization

There is growing use of assessments of animal vocalizations to understand animal models of neuropsychiatric illness and speech disorders. Critically, the computational tools used to analyze these vocalizations are cumbersome and analyses of vocalizations often depend upon highly arbitrary classification schemes. Through a collaboration between an expert in rodent social behavior (Dr. Garet Lahvis) and experts in signal processing of human speech (Drs. Kain and van Santen), the project will develop software that automatically segments, tracks, and classifies mouse USVs from acoustic recordings of rodents. This system, which can speed up data collection by orders of magnitude, will have widespread application for academia and industry for pharmacological, toxicological and genetic research.

September, 2013

BioSpeech Inc. receives $234,000 for new NIH SBIR Phase I Sleep Apnea project.

Sleep disordered breathing (SDB) is believed to be a widespread, underdiagnosed condition associated with detrimental health problems, at a high cost to society. The current gold standard for diagnosis of SDB is a time consuming, expensive, and obtrusive (requiring many attached wires) sleep study, or polysomnography (PSG).

The proposed work will create a portable, low-cost, and minimally obtrusive system for automatically detecting sleep-disordered breathing, such as cessation of breathing (apnea). The system will enable early and frequent screening for sleep-disordered breathing in a patient’s home, significantly increasing patient comfort while capturing more representative sleep data compared to a clinical sleep study requiring many attached wires.

This grant strengthens our mission of detecting health conditions through computational analysis of behavioral signs.