摘要 |
PROBLEM TO BE SOLVED: To inform a user of the out-of-vocabulary when the speech signal input with a speech recognition request is out of vocabulary. SOLUTION: This speech recognizing device informs the user of the out-of-vocabulary when the speech signal input with the speech recognition request is out of vocabulary. The speech recognizing device comprises a recognition processing request receiving means 10 for receiving input of a speech recognition request and a speech signal from the user, an out-of-vocabulary determining means 30 for determining whether the input speech signal is out of vocabulary, and a speech recognition processing means 20 for performing speech recognition processing of the input speech signal using audio model and language model. When the input speech signal is out of vocabulary, this speech recognizing device informs the user of out-of-vocabulary. COPYRIGHT: (C)2010,JPO&INPIT |