The objectives of the proposed research are to quantify and classify vocal disorders caused by variations in laryngeal and oral tract function that may be brought about by a structural alteration to the speech production system by a pathology or changes in physiology. The proposed methodology for the research is to model vocal disorders with speech synthesizers that include both phonatory and resonance features, each capable of being separately controlled. Laboratory speech synthesizers are now able to generate exceptionally high quality, natural sounding speech through finely controlled software algorithms that may vary speech production features precisely in degrees, one at a time, to synthesize vocal disorders. This methodology is also able to provide auditory scientists and clinicians nev test signals that may be used to study auditory disorders. These test signals can be precisely controlled, repeated without variation and designed to the specification of the auditory researcher. The expected results would enhance the clinician's ability to identify, quantify, and rank order the most likely laryngeal (phonatory) and oral tract (resonance) characteristics that contribute to a specific vocal disorder, e.g., hoarseness. The study will model cause-and-effect relationships between voice production features and the acoustic signal. These relationships may lead to more quantitative descriptors (terminology) of voice disorders that might eventually be universal. The research is not directed to the diagnosis of disease based on acoustic factors, rather the purpose is to establish quantitative descriptors of the relationships between symptoms, pathology, physiology, and speech production characteristics.
