Browse Prior Art Database

Word Translation Profiles for Computer Input from a Speech Recognition System

IP.com Disclosure Number: IPCOM000104149D
Original Publication Date: 1993-Mar-01
Included in the Prior Art Database: 2005-Mar-18
Document File: 2 page(s) / 57K

Publishing Venue

IBM

Related People

Andreshak, JC: AUTHOR [+4]

Abstract

Disclosed is a method of mapping spoken words into different text or command actions for different target applications.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Word Translation Profiles for Computer Input from a Speech Recognition System

      Disclosed is a method of mapping spoken words into different
text or command actions for different target applications.

      When  recognized words from a speech recognition system are fed
as text or command input into computer applications, the text or
command which is generated from a given spoken word or phrase is
dependent on which application is receiving input.  Different
computer applications have different user interfaces and different
sets of input commands or text which are required to use the
application.

      The translation of speech into text or commands cannot be fixed
if more than one application is to be used with speech input, but
must have the ability to dynamically change as different applications
are run by the user.  Switching rapidly between many different
applications is increasingly important with the advent of windowing
systems such as X-Windows on UNIX* and Presentation Manager on
OS/2**.

      In addition, the sets of words or commands which are valid
input to a given computer application may change depending on the
state of the application.  A word processor program, for example, may
use different sets of menus or commands depending on which set of
functions is valid at any given time during the creation of a
document.  If an application is to receive input from  (i.e., be
"driven by") a speech recognition system, then the output of the
recog...