Biomaterial Database

Novel approach to acoustical voice analysis using artificial neural networks. 2000

R Schönweiler, and M Hess, and P Wübbelt, and M Ptok

Department of Communication Disorders, Center of Ophthalmology, Otorhinolaryngology and Communication Disorders, Hannover Medical School, Germany. schoenweiler.rainer@mh-hannover.de

Perceptual rating scales are widely used for the assessment of voice quality. These ratings may be influenced by the individual experience of the listener. Thus, researchers have turned to acoustical measures which may eventually correlate with voice quality. In this study we tested whether multivariate statistics, combined with artificial neural networks, could identify patterns of acoustic voice parameters corresponding to a widely used perceptual rating scale. In a multicenter study with 31 raters, voice samples of 117 individuals with or without voice disorders were perceptually rated. The RBH index, consisting of a 4-point scale of roughness, breathiness, and hoarseness, was used. Voice samples were then analyzed with an acoustical feature extraction and classified using amultivariate regression tree analysis with the perceptual ratings as a priori information. Artificial neural networks were trained to selected acoustic parameters having high "relative importance" in the regression trees. Mean classification accuracies were around 30% with topographic feature maps (trained with Learning Vector Quantization algorithm) and 65-85% with feedforward networks (trained with RProp algorithm). Based on the best-fitting results with feedforward networks, a classification system (computer program) consisting of 50 simultaneous working networks was developed. Using this program, the classification matched 40% of the a prori values in both R and B domains. In 65% they matched at least in one domain. These accuracies are within the range reported by other authors using artificial neural networks in biology and clinical medicine. Thus, the results encourage further research of feedforward networks for acoustic voice analysis.

UI	MeSH Term	Description	Entries
D006801	Humans	Members of the species Homo sapiens.	Homo sapiens,Man (Taxonomy),Human,Man, Modern,Modern Man
D013061	Speech Acoustics	The acoustic aspects of speech in terms of frequency, intensity, and time.	Acoustics, Speech,Acoustic, Speech,Speech Acoustic
D014833	Voice Quality	That component of SPEECH which gives the primary distinction to a given speaker's VOICE when pitch and loudness are excluded. It involves both phonatory and resonatory characteristics. Some of the descriptions of voice quality are harshness, breathiness and nasality.	Qualities, Voice,Quality, Voice,Voice Qualities
D016571	Neural Networks, Computer	A computer architecture, implementable in either hardware or software, modeled after biological neural networks. Like the biological system in which the processing capability is a result of the interconnection strengths between arrays of nonlinear processing nodes, computerized neural networks, often called perceptrons or multilayer connectionist models, consist of neuron-like units. A homogeneous group of units makes up a layer. These networks are good at pattern recognition. They are adaptive, performing tasks by example, and thus are better for decision-making than are linear learning machines or cluster analysis. They do not require explicit programming.	Computational Neural Networks,Connectionist Models,Models, Neural Network,Neural Network Models,Neural Networks (Computer),Perceptrons,Computational Neural Network,Computer Neural Network,Computer Neural Networks,Connectionist Model,Model, Connectionist,Model, Neural Network,Models, Connectionist,Network Model, Neural,Network Models, Neural,Network, Computational Neural,Network, Computer Neural,Network, Neural (Computer),Networks, Computational Neural,Networks, Computer Neural,Networks, Neural (Computer),Neural Network (Computer),Neural Network Model,Neural Network, Computational,Neural Network, Computer,Neural Networks, Computational,Perceptron