Abstract
Background: Voice is one aspect of speech. In this study, evaluation of the audio signal is used to determine the severity of the speech disorder and intelligibility. Until now, methods studied the quality of speech.
Methods: In consultation with the specialists a sentence with 5 words and 8 syllables has been determined to test 80 sample random persons were randomly selected among healthy people with different age groups and gender. Unhealthy people randomly selected from among the pupils centers. Among the different grading unhealthy and speech disorders were present. The cases were 50 healthy people and 30 patients with speech disorders. Using the model proposed in this paper, an intelligent system designed base on the principles of speech and image processing for detection speech disorder.
Conclusion: The proposed system has been tested on the different voice signals and statistical data have been extracted. The results represent a breakdown of people with an accuracy of more than 92% has been compared with the results of other reports, the result is very acceptable. The percentage of the proposed system error that leads to a repeat of the experiment is also about 7%.