In the recent decades a lot of researches focused on the computer based human speech processing and speech recognition with varying success. There are two core elements of such systems. The first one is the feature vector of the given voice containing its most relevant parameters. The other one is an artificial intelligence (AI) system, which is able to distinguish between different cases and make decisions in specific situations.
During my researche my goal is to find such nonlinear parameters, which can be used to determine the disease using only the patient’s voice and which is good enough for teaching an AI system. I have approached the complexity of the speech producing as a chaotic system and I have used one of the most characteristic parameters of chaos, the fractal dimension (Df).
Using these results and variables, I have designed a database for the further speech processing, that identifies 78 different parameters according to the vowels “e” and “o”. During my thesis work I have used different statistical approaches and methods like principal component analysis (PCA), two- and multidimensional linear discriminate analysis (LDA), linear and non-liner regression analysis, logistic regression, support vector machines (SVM), support vector regression (SVR) in order to analyze the characteristics of the data set. I have used the results to define the best and optimal feature vectors which are significant to differentiate the healthy and unhealthy voices, quality of the voices according to the RBH scale or good enough to identify the pathological problems behind the voice disorders. During the classification measurements and comparisaions these feature vectors reached higher accuracy than previously used ones.
Within my thesis I have also worked on the concept of developing an infocommunication based system, that relies on eHelath and Telemedicine methodologies, in order to increase the quality of the diagnoses of speech disorders and improve the quality of the speech therapies. A system like this would provide a fast and easy connection between patients and doctors to help the diagnosis and the monitoring of the patients. It would also provide easy adjustment options for the speech therapies and provide help to teach the patients the correct voice production without leaving their homes.