Speech visualization system based on integrating feature for the hearing handicapped

This paper described a new speech visualization system that created readable patterns by integrating different speech features into a single image. The system extracted features from speech signals and converted them into a visual image. The method remedied the defect that single feature made the deaf accept difficultly because of too professional. The auditory feature was displayed on the CRT by plot patterns and the deaf could utilize their own brain to identify different speech via training their oral ability effectively. The system was comprised of three main sections, a pre-processing section, a feature extracting section, and speech plot display section. We evaluates the visual system in a preliminary test, The simulation results show that the approach is correct and effective, and the correct rate reaches 98.46% for vowel and diphthong.