Browse Prior Art Database

On-Line Distinction between Text Editing and Speech Recognition Adaption

IP.com Disclosure Number: IPCOM000113914D
Original Publication Date: 1994-Oct-01
Included in the Prior Art Database: 2005-Mar-27
Document File: 2 page(s) / 51K

Publishing Venue

IBM

Related People

Goldwyn, RM: AUTHOR [+3]

Abstract

Disclosed is a feature for a dictation system having text entry by speech and by keyboard, through which a user reviewing a document can make changes to the document by keyboard, while indicating to the system which changes are required due to a misinterpretation by the speech recognition process within the system. Using this information, the system adapts its operation to the speech of the individual user.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 62% of the total text.

On-Line Distinction between Text Editing and Speech Recognition Adaption

      Disclosed is a feature for a dictation system having text entry
by speech and by keyboard, through which a user reviewing a document
can make changes to the document by keyboard, while indicating to the
system which changes are required due to a misinterpretation by the
speech recognition process within the system.  Using this
information, the system adapts its operation to the speech of the
individual user.

      During the process of editing a document developed with a
system of this kind, some changes will typically be made due to
misinterpretations by the speech recognition process, while other
changes will be made in a process of traditional text editing.  The
distinction between these types of changes is important because, if
speech recognition adaption is applied when only text editing is
intended, future speech recognition suffers, with changes intended
for only one document propagating to all future dictation, depending
on the acoustic similarity of the dictated material.  On the other
hand, if only text editing is applied to correct mistakes made by the
speech recognition feature, this feature does not have a chance to
capture parameters which would allow it to adapt to the individual
speaker.

      Thus, the system presents the user with the message, "From now
on, when you say xxx, yyy will by typed.  OK?" where xxx is an audio
playback of the target word and yyy is t...