摘要 |
<p>PROBLEM TO BE SOLVED: To perform smooth communication between a hearing-impaired person and a non-impaired person. SOLUTION: Voice data from the non-impaired person are converted into a character string showing the contents of speech of the non-impaired person by a voice recognizing part 11 and afterwards, speech data from which a redundant way of speech is removed, are prepared by a noise removing part 12. A control part 14 applies these speech data to a display processing part 15 which displays these data, and applies the same speech data to a predictive data learning part 16 which prepares predictive data for predicting the character string of a final form inputted by the hearing-impaired person from the speech data. The prepared predictive data are registered in a predictive database 18 and used when the hearing-impaired person inputs characters.</p> |