vocal fold disorder diagnosis
Loading...
Date
item.page.authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Speech disorder detection is of constant importance in our lives today. In
newlinemodern society, the problem of voice dysfunction caused by various routine activities
newlinehas become an enormous concern. This impairment is often manifested in voice signals
newlinein terms of factors of sound perception like harshness, breathiness, hoarseness and
newlinesimilar to this.
newlineOne of the perspectives for the categorization of a voice disorder is a
newlinepsychogenic disorder where poor voice quality becomes symbolic or severe due to
newlinemany psychological issues like stress as well as emotions. Mainly emotions like
newlinehappiness, anger, surprise, sadness, and calm behaviour make much of an impact on the
newlinespeaking behavior of the person. The voice disorder with organic (structural and
newlineneurogenic) and functional (physical structure variation) has some limitations based on
newlinethe boundless scope of discrepancy and deviation during the assessment. The person
newlinewhen becomes emotionally unstable, which impacts on his/ her voice production
newlinesystem, which further causes stuttering and stammering behavior while speaking. To
newlineovercome the limitations of the other two types of voice disorders mentioned above,
newlinethe deduction of voice pathologies in the context of emotions in the form of
newlineperformance parameters is very important and beneficial for improving the overall life
newlineof a person.
newlineThe basic objective is to focus on the development of a tool with an
newlineacoustoglottography as an effective voice sample phenomenon in the preprocessing
newlinealong with a new approach of mathematical modelling of finding residue of the
newlineextracted signal with a fusion of numerous features with a unique summation method
newlinenamed DAFF (Dysfunction and Affective Feature Fusion) which forms a multimodal
newlinefeature fusion set for the improvement of recognition accuracy. The features are
newlinedistributed based on the time and frequency variations as high-level and low-level based
newlineon prosodic and spectral features. Zero Crossing Rate, Energy parameter with its
newlineEntropy describing the pathology and emotions justif