| 摘要 |
PROBLEM TO BE SOLVED: To provide a speech recognition device with high speech recognition performance. SOLUTION: The speech recognition device has a speech input means for inputting speech which a user utters, a speech recognition means for recognizing the speech, a language dictionary means for storing a language dictionary which describes vocabulary to be a recognition object of the speech recognition means, an image pickup means for picking up video images around the user, a discrimination information extraction means for extracting a discrimination information image among the video images and converting them into discrimination information character strings, a next utterance prediction means for determining candidate vocabulary (next utterance candidate vocabulary) of speech which the user utters next using the discrimination information character strings and a language dictionary changing means for changing the language dictionary using the next utterance candidate vocabulary. COPYRIGHT: (C)2005,JPO&NCIPI |
