| 摘要 |
PROBLEM TO BE SOLVED: To provide an apparatus and a method for varying speaking speed that make the contents of a reproduced speech easy to understand even when the reproduction speed is varied or when the reproduction time is shortened. SOLUTION: A speech data generation part 101 converts a speech signal from a microphone 100 into speech data. A text data generation part 102 analyzes the speech contents of the speech data for conversion into text data. A unit division part 103a divides the text data into units having linguistic meanings. An input part 104 receives reproduction magnification N and a total reproduction time L as information regarding the speed of speech reproduction. A speech reproduction control part 105a calculates reproduction times for each divided units based on the data inputted from the unit division part 103a and data inputted from the input part 104, converts the speech data inputted from the speech data generation part 101 into speech signals for each unit based on the calculated reproduction time, and outputs the speech signals from a loudspeaker 106. COPYRIGHT: (C)2004,JPO&NCIPI
|
