Skip to Main Content
The amount of non-invasive methods of diagnosis has increased due to the need for simple, quick and painless tests. Due to the growth of technology that provides the means for extraction and signal processing, new analytical methods have been developed to understand the complexity of the voice signals. This paper presents a new idea to characterize signals of healthy and pathological voice based on two mathematical tools widely known in the literature, Wavelet Transform (WT) and Artificial Neural Networks. Four classes of samples were used: one from healthy individuals and three from people with vocal fold nodules, Reinke's edema and neurological dysphonia. All the samples were recorded using the vowel /a/ in Brazilian Portuguese. The work shows that the proposed approach using WT is a suitable technique to discriminate between healthy and pathological voices.