首页> 外国专利> INTERACTIVE DATABASE RETRIEVING METHOD AND DEVICE AND RECORDING MEDIUM RECORDED WITH INTERACTIVE DATABASE RETRIEVING PROGRAM

INTERACTIVE DATABASE RETRIEVING METHOD AND DEVICE AND RECORDING MEDIUM RECORDED WITH INTERACTIVE DATABASE RETRIEVING PROGRAM

机译:交互式数据库检索方法和装置以及使用交互式数据库检索程序记录的介质

摘要

PROBLEM TO BE SOLVED: To equivalently enhance the recognizing likelihood of voice retrieval keys. ;SOLUTION: Likelihood (retrieval key recognizing likelihood) with respect to respective retrieval keys are calculated by voice recognizing and processings of voice retrieval keys while using a retrieval database (S2) and when retrieval keys whose likelihoods are equal to or larger than a threshold value are equal to or smaller than a prescribed number, a correct/error property is questioned as to retrieval keys whose likelihoods are equal to or larger than the threshold value and inquiries for determining retrieval keys are presented by voices (S5). Then, when retrieval keys whose likelihoods are equal to or larger than the threshold value are equal to or larger than the prescribe number or zero, related inquiries for determining retrieval keys are present by voices (S9) on the basis of related information database holding the attribute of the retrieval keys (for example, names of the administrative division of Japan as the attribute of names of cities, towns and villages) as related information. Related information and its recognizing likelihood (related information recognizing likelihood) are calculated by voice recognizing processings to the responces and a retrieval key recognizing likelihood and the related information recognizing likelihood are integrated (S11) and inquiries questioning the correct/error property are performed as to the retrieval keys whose likelihood are equal to or larger than the threshold value of the integrated recognizing likelihood on the basis of the related information database.;COPYRIGHT: (C)2000,JPO
机译:要解决的问题:同等地提高语音检索键的识别可能性。 ;解决方案:在使用检索数据库(S2)时以及当似然性等于或大于阈值的检索关键码时,通过语音识别和语音检索关键码的处理来计算各个检索关键码的似然度(检索关键码识别可能性)如果等于或小于规定数量,则对似然性等于或大于阈值的检索关键字询问正确/错误属性,并且通过语音呈现用于确定检索关键字的查询(S5)。然后,当似然性等于或大于阈值的检索关键字等于或大于处方数或零时,基于保持该关键字的相关信息数据库,由语音呈现用于确定检索关键字的相关查询(S9)。检索关键字的属性(例如,日本行政区划的名称作为城市,乡镇名称的属性)作为相关信息。通过响应于语音的语音识别处理来计算相关信息及其识别可能性(相关信息识别可能性),并且将检索关键字识别可能性和相关信息识别可能性进行整合(S11),并且执行关于正确/错误特性的询问。基于相关信息数据库的似然度等于或大于综合识别似然度阈值的检索关键字。版权所有:(C)2000,JPO

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号