摘要 |
PROBLEM TO BE SOLVED: To provide a speech recognition system capable of accurately detecting user's incorrect operation by reducing processing load for detecting an utterance section. SOLUTION: The speech recognition system includes: an utterance timing indication obtaining means for obtaining user's indication of utterance timing including indication of utterance start; a speech signal holding means which holds an input speech signal, and which outputs the held speech signal and speech signals which are input after that, when indication of utterance speech is obtained by the utterance timing indication obtaining means; an utterance section detecting means for detecting the utterance section from the speech signal output by the speech signal holding means; and an incorrect operation detecting means for detecting user's incorrect operation based on the utterance section detected by the utterance section detecting means, and indication of utterance timing obtained by the utterance estimating indication obtaining means. COPYRIGHT: (C)2011,JPO&INPIT |