Acoustic analysis of voice for computerized laryngeal pathology assessment

The purpose of this paper is to review the techniques of computerized acoustic analysis of voice signals used for laryngeal pathology assessment. Results of using a methodology described by Davis for distinguishing between T1 glottic cancer, noncancerous pathology, and normal speakers are presented. These results indicate that this technique is not sufficiently sensitive for general clinical usefulness. There is an intentional lack of mathematical detail in this paper, as its prime intent is to conceptually outline issues and approaches.