(东北大学 信息科学与工程学院, 沈阳 110004)
Speech visualization based on wavelet transform
WANG Xu, XUE Lifang, YANG Dan, HAN Zhiyan
(College of Information Science & Engineering, Northeastern University, Shenyang 110004, China)
Abstract:This paper described a new speech visualization method that created readable patterns by integrating combined feature into a single image. The system made use of timefrequency analysis based on wavelet transform to simulate the bandpass filter property of basilar membrane. The method remedied the defect that short fourier transform(SFT) had the same timeresolution and frequencyresolution to different frequency ranges. The auditory feature was displayed on the CRT by plot patterns and the deaf could utilize their own brain to identify different speech for training their oral ability effectively. Firstly, speech signal underwent a series of preprocessing course. Secondly, made use of wavelet transform to process timefrequency analysis for speech signal and extracted the feature value for speech visualization. Then calculated that the feature value lay in which place in full array and obtained the combined feature value. Finally, utilized plot display algorithm to generate a speech plot. ......