The goal of the paper is to present a speech nonfluency detection method based on linear prediction coefficients obtained by using the covariance method. The application “Dabar” was created for research. It implements three different methods of LP with the ability to send coefficients computed by them into the input of Kohonen networks. Neural networks were used to classify utterances in categories of fluent and nonfluent. The first one was Kohonen network (SOM), used to reduce LP coefficients representation of each window, which were used as input data to SOM input layer, to a vector of winning neurons of SOM output layer. Radial Basis Function (RBF) networks, linear networks and Multi-Layer Perceptrons were used as classifiers. The research was based on 55 fluent samples and 54 samples with blockades on plosives (p, b, d, t, k, g). The examination was finished with the outcome of 76% classifying.
2
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
The therapy of stuttering people is a time-consuming and long-Iasting process which requires a great effort both from the logopaedist and patient. The process can be divided into three parts: recording of patient's utterances (reading, telling, conversation), 20-minute corrective exercises with the echo (reading, tell ing) and individual work of the stuttering person with difficult words. All of these tasks may be performed with the use of a computer, controlled by a special program elaborated for that purpose. The computer system for the logopaedic diagnosis and therapy (DTL) allows for recording and saving utterances as sound files, practice with acoustical or visual echo and performance of automatically generated tasks adjusted to individual difficulties of particular speakers. Examples of analyses performed at various periods of therapy, i.e. at the beginning, during and after the therapy, supply information conceming e.g. the stuttering intensity and types of the occurring errors. The results presented in this work concern the control recordings performed at 1-1.5-month periods of time for twelve patients.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.