Compare two methods to measure consonant perception in noise

A common problem with measuring consonant perception in noise is the unpredictable audibility of each consonant (e.g., given the same signal level, some consonants are well audible, while others are not). The purpose of this project is to examine two methods in speech perception to solve this problem and also to evaluate their time efficiency and data reliability. These two methods, frequently used in psychological measures, are method of limit and method of constant stimuli. Preliminary results showed that the method of limit was more time efficient while both methods provided quite reliable data. The advantages and disadvantages of the two methods will be discussed.