Using patient's speech signal for vocal ford disorders detection based on lifting scheme

摘要

Regarding to the impress of speech in community relations establishment and the effect of the larynx in speech, correct and timely diagnosis of diseases of vocal cords have particular importance. Since the Conventional methods for diagnosis of vocal cords are usually slow, expensive and annoying, so the purpose of this paper is to analysis and classify of vocal fold disorders with the help of audio signal processing vowel /a/. This non-invasive method is cheaper, fast and repeatable. The database used for this work was developed by Massachusetts Eye and Ear Infirmary (MEEI) voice and speech. Although common wavelet features have acceptable performance, but expected that design optimization features of adaptive wavelet based on lifting method lead to improve results. To design the adaptive wavelet transform, the parameters of lifting scheme generating biorthogonal wavelet are initially applied and then they are optimized through genetic algorithm and classification performance of support vector machine. The result separation of normal and pathological signals provides an accuracy of 98.30%. Also, the result of two-class separation based on lifting scheme indicative the advantage of this suggested method with other wavelets.