| Abstrakt: |
Stuttering is a speech disorder, causes frequent problem with normal flow of speech. The normal fluency of the speech is disturbed by uncontrolled prolongation and repetitions of words, sounds, phrases or syllables and unwanted blockages in speech. Researchers are more interested in this field because in the field of speech language pathology and speech signal analysis, stuttering plays an important role. Stutter speech processing found its application in stuttering assessment and analysis that helps a speech language pathologist in the treatment and diagnosis. Thus, understanding of the stutter speech behavior by the researchers is a fundamental requirement and that helps to grasp the subject of interest and problem solving. Hence keeping this object in mind, this paper provides an insight to overview of the conventional assessment techniques and objective assessment techniques used for stuttering assessment. It's also provides insight to the further development in this area. The paper summarizes the different stuttering disfluency classification techniques and introduces the subjective and objective assessment of stuttering and its applications till current date. Summary of the classifiers table included at the end of the paper for better understanding. In addition, the previously used classification techniques are explained to make the paper self-sufficient and they are Artificial Neural Networks (ANN), Hidden Markov Model (HMM), k-nearest neighbor (k-NN), Linear Discriminate Analysis (LDA), and Support Vector Machine (SVM). [ABSTRACT FROM AUTHOR] |
