Discriminative and Maximum Likelihood Classifiers for Computer-Based Visual Feedback for Speech Training for the Hearing Impaired

A visual speech training aid for persons with hearing impairments has been developed using a Windows-based multimedia computer. The training aid provides real time visual feedback as to the quality of pronunciation for 10 steady-state American English monopthong vowel phonemes (/aa/, /iy/, /uw/, /ae/, /er/, /ih/, /eh/, /ao/, /ah/, and /uh/). This training aid is thus referred to as a Vowel Articulation Training Aid (VATA). Neural network (NN) classifiers are used to produce the two main displays: a 10-category “vowel bargraph” which provides “discrete” feedback, and an “ellipse display” which provides continuous feedback over a 2-D field, similar to a formant1formant2 display often used by speech scientists to depict vowels. Continuous feedback such as this is desirable for speech training to help improve articulation. In this paper a method is described for combining the neural network classifiers used in the speech display with maximum likelihood classifiers which help to reject those sounds which were not used for training the neural network.