摘要 |
PROBLEM TO BE SOLVED: To provide a system and method for speech recognition capable of looking down at a result of a speech recognition to two or more utterances included in a comparatively long input speech data, and to provide a program therefor. SOLUTION: The system for speech recognition comprises: a speech input means 110 for inputting a speech data; a speech recognition means 120 for receiving the input speech data from the speech input means to perform speech recognition; a speech recognition evaluating means 140 for obtaining a priority to a result of the speech recognition by the speech recognition means and/or each part constituting the recognition result; a storage means 130 for storing the recognition result and the priority; a recognition result rectifying means 150 for judging display/non-display of the recognition result and/or each part constituting the recognition result following the priority, and creating output information; and an output means 160 for outputting the output information. COPYRIGHT: (C)2007,JPO&INPIT
|