US 5,909,667 AGranted

Method and apparatus for fast voice selection of error words in dictated text

AssigneeInternational Business Machines Corporation

InventorsKyriakos Leontiades, Kris Coe

Life of this patent

Estimating the term…

Loading patent…

Abstract

A computer voice operated dictations system having a text editing mode provided for editing dictated text. The text editing mode navigates a user through the dictated text by highlighting a correction window comprised of a plurality of words comprising a portion of the text. The correction window is comprised of "m" words and the value of m is preferably determined, based upon the voice recognition capabilities of a particular voice recognition engine associated with the dictation system. The greater the ability of the speech recognition engine to recognize and distinguish spoken words, the larger the value of m. A correction vocabulary is defined which consists of the plurality of words which are contained within the correction window and a plurality of correction commands. The voice recognition computer dictation system is configured for processing a spoken utterance which corresponds to an error word contained within the plurality of words highlighted in the correction window is processed. When the error word is spoken by a user it uniquely identifies the word which is to be edited. Thereafter, the error word can be immediately corrected by the user articulating a further voice command.

Claims27

Dependent claims

In a computer system (20) having an operating system (24), a memory (27), and processing means (26, 28) for receiving spoken words and converting said spoken words to words for storage in said memory (27) and for display on a visual display (32), a method for voice-activated correction of text, said method comprising:

(a) storing in said memory (27) a plurality of words (60) corresponding to said spoken words which are enabled for correction by voice activation, alternate word lists containing alternate words which are also likely matches for each spoken word, and a plurality of correction commands;
(b) displaying one or more of the words (60) enabled for correction in a first demarcated correction region (130) on the visual display (32) in response to one of said plurality of correction commands being spoken by a user, said demarcated correction region identifying words which are available for a user to say and correct;
(c) receiving and identifying in the processing means (26, 28), an error word (132) spoken by a user and matching said spoken error word to a word which exists in said demarcated correction region;
(d) operating said processing means (26, 28) to retrieve from said memory (27) and to display on said visual display (32), one or more alternate words (136) from said alternate word list for said error word as potential replacements for said error word (132);
(e) receiving and identifying in the processing means (26, 28) a second one of said plurality of correction commands, in response to said second one of said plurality of correction commands (136, 138) being spoken by a user;
and
wherein the processing means (26, 28) is responsive to said second one of said plurality of correction commands to correct the error word on the visual display.

4 more elements

In a computer system (20) having an operating system (24), a computer memory (27), and a user interface display (32), an apparatus for voice-activated correction of text, said apparatus comprising:

(a) means for storing in said memory (27) a plurality of words (60) corresponding to words spoken by a user which are enabled for correction by voice activation, alternate word lists containing alternate words which are also likely matches for each spoken word, and a plurality of voice activated correction commands;
(b) means for displaying one or more of the words (60) enabled for correction in a first demarcated correction region (130) on the visual display (32), said demarcated correction region identifying words which are available for a user to say and correct;
(c) means for receiving and identifying an error word (132) to be corrected when said error word is spoken by a user, said error word (132) being contained within said demarcated correction region, and matching said word to a word which exists in said demarcated correction region;
(d) means for retrieving from said memory and displaying on said visual display (32) one or more alternate words (136) from said alternate word list for said error word as potential replacements for said error word;
(e) means responsive to one of said plurality of correction commands being spoken by a user, for correcting the error word on the visual display.

2 more elements

In a computer system having an operating system, a memory, and processing means for receiving spoken words and converting spoken words to text words for storage in memory and display on a visual display, an apparatus for voice-activated correction of text, said apparatus comprising:

(a) text processing means for
(1) selecting a first set of correction words enabled for voice activated correction from the text words stored in said memory,
(2) for enabling a first group of voice activated correction commands, and
(3) for displaying the first set of correction words in a first demarcated correction region on the visual display;
(b) means for receiving and storing in said memory (27) an error word (132), when said error word (132) is spoken by a user, and for communicating said error word (132) to said text processing means;
(c) means for matching said error word (132) with one of said correction words;
(d) means for displaying a list of alternate words (64) on the visual display (32) of the computer system (20) in response to said error word (132) being spoken, said alternate word list created for each word spoken during a dictation mode, said alternate word list containing alternate words which are also likely matches for said spoken error word;
(e) means for receiving and storing a spoken correction command (136, 138) in the computer system to replace said error word with a word contained on said alternate word list;
whereby said computer system is responsive to the spoken correction command to correct the error word on the visual display.

6 more elements

Dependencies are inferred from the claim text.