发明名称 |
DEVICE AND METHOD FOR SPEECH RECOGNITION, DEVICE AND METHOD FOR NAVIGATION, AND AUTOMOBILE |
摘要 |
PROBLEM TO BE SOLVED: To easily operate at high level a navigation device by excluding a recognition object word of a last recognized speech and performing a speech recognizing process when the same aural signal is inputted. within a specified time again. SOLUTION: A history list in a speech recognizing circuit 14 is not cleared unless a specific time (10 seconds) has lapsed after last speaking. Under the control of the speech recognizing circuit 14, the inputted speech is recognized. When the history list contains te same data as speech data on a candidate obtained by the recognition, the data is deleted from recognized candidates. Then, the data having the highest recognition degree (matching degree) among data on the remaining candidates is supplied as a recognition result to the speech synthesizing circuit 31 of the navigation device 20 and outputted as a voice from a speaker 32. Consequently, a wrong place name is prevented from being recognized again even if the place name is spoken again. |
申请公布号 |
JPH09114487(A) |
申请公布日期 |
1997.05.02 |
申请号 |
JP19950267541 |
申请日期 |
1995.10.16 |
申请人 |
SONY CORP |
发明人 |
ISHII KAZUO;YAMAMOTO EIJI;TANAKA MIYUKI;TSUNODA HIROSHI;ASANO KOJI;OGAWA HIROAKI;OMOTE MASANORI;MINAMINO KATSUKI |
分类号 |
G09B29/10;G01C21/00;G08G1/0969;G10L13/00;G10L15/00;G10L15/22;G10L15/28;(IPC1-7):G10L3/00;G10L3/00;G08G1/096 |
主分类号 |
G09B29/10 |
代理机构 |
|
代理人 |
|
主权项 |
|
地址 |
|