Voice analysis technology before and after neck surgery
摘要
Monitoring and prevention of speech dysfunction after surgery on organs in the neck area (larynx, thyroid gland, trachea, and esophagus) is a pressing issue at the intersection of medicine and information technology. Thie present article proposes a technology for voice analysis before and after surgery based on the empirical mode decomposition method and formation of a complex voice signal. A database of 1000 speech signals recorded in 21 patients before and after surgery was compiled. The study results showed that eight out of 18 women and one out of three men experienced increases in the frequency of the fundamental tone, while the remaining patients experienced decreases in the frequency range. The significance of the results was confirmed by visualization of the presence or absence of surgical injury to the recurrent laryngeal nerve in the study patients.