Though "typical" patterns can be abstracted from speech sounds, the actual patterns vary widely. The human mind extracts many patterns, making its final selection on the basis of meaning. Until computers are capable of semantic analysis, voice-operated equipment must rely on "constrained" speech, which restricts itself to a simply abstracted pattern. A vowel meter has been built which abstracts such a pattern from vowel sounds. With this equipment and others of a similar nature, a training program will be undertaken to determine the effects of training upon speech precision, in the hope that such training will make possible a voice-operated brailler for communication with the deaf-blind. The underlying principles of the vowel meter are explained.
No takes yet. Share an insight, caveat, or question.
Milton Rosenstein (1965) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: