-
公开(公告)号:JP2000284795A
公开(公告)日:2000-10-13
申请号:JP2000061542
申请日:2000-03-07
Applicant: IBM
Inventor: JAMES R LEWIS , BALLARD BARBARA
Abstract: PROBLEM TO BE SOLVED: To improve the total accuracy of voice recognizing processing related to a user by providing a step for monitoring an application program and a step for discriminating whether or not an erroneously recognized word in a test dictating one text correction input is to be corrected. SOLUTION: An operating system 24 and an audio text processor application 28 are monitored and all the changes or corrections executed by the user are identified (S52). Next, it is first discriminated whether or not a certain dictated text is displayed while being intensified or selected by the user (S54). In the case of YES, it is secondly discriminated whether or not a new text is inserted in the text displayed while being intensified by the user (S56). Then, it is discriminated whether the text displayed while being intensified is deleted or not (S58), and after cursor confirmation, it is discriminated whether the new text can be inserted in place of the deleted text or not (S62 and S64).
-
公开(公告)号:GB2347774A
公开(公告)日:2000-09-13
申请号:GB0002504
申请日:2000-02-04
Applicant: IBM
Inventor: LEWIS JAMES R , BALLARD BARBARA
Abstract: In a computer speech dictation system, a decision is made whether to update a language model based upon user amendments to text in a dictated word processing document. A user's text-amending input is examined to determine whether the text-amending input corrects a misrecogized word in the dictated text. If the text amending input corrects a misrecognized word, the system updates a language model based upon the text which has been amended.
-
公开(公告)号:GB2347774B
公开(公告)日:2003-06-25
申请号:GB0002504
申请日:2000-02-04
Applicant: IBM
Inventor: LEWIS JAMES R , BALLARD BARBARA
-
公开(公告)号:CA2297414A1
公开(公告)日:2000-09-08
申请号:CA2297414
申请日:2000-01-28
Applicant: IBM
Inventor: LEWIS JAMES R , BALLARD BARBARA
Abstract: In a computer speech dictation system, a method and system for automatically determining whether to update a language model based upon user amendments to text in a dictated word processing document. The method includes the steps of monitoring a user's textamending input to an application program for amending the dictated text and subsequently performing processing responsive to the monitoring step to determine whether the text-amending input corrects a misrecognized word in the dictated text. If the test amending input corrects a misrecognized word, the system updates a language model based upon the text which has been amended.
-
-
-