Voice pathology distinction using autoassociative neural networks

Acoustic analysis is a non-invasive technique that supports voice disease screening, especially the detection and diagnosis of distinction between chosen voice pathologies and healthy control group. This work put en effort on creation of efficient and accurate system for automatic detection and differentiation of normal and three different voice pathologies. This system ensures non-invasive and fully automated analysis of voice characteristics and decision system based on neural networks. The feature vector describing the vocal tract is set up from 35 parameters. Recordings of patients suffering from hyperfunctional dysphonia, recurrent laryngeal nerve paralysis, laryngitis and healthy control group are considered in our experiments. From the experimental results it is observed that effectiveness of auto-associative neural networks seems to be promising in the application of pathological voice distinction.