Word processor via voice

Details Word processor via voice Word processor via voice

1998-11-24

2001-04-03

{haeck over (S)}mits, T{overscore (a)}livaldis I. (Department: 2741)

Data processing: speech signal processing, linguistics, language

Speech signal processing

Recognition

C704S275000, C704S231000, C704S270000, C382S229000

Reexamination Certificate

active

06212497

ABSTRACT:

FIELD OF THE INVENTION
The present invention relates to a word processor, and in particular to a word processor for inputting words by way of voice.
This application is based on Japanese Patent Application No. Hei 9-323594, the contents of which are incorporated herein by reference.
BACKGROUND ART
Various voice recognition apparatuses for inputting words by way of the voice have been proposed. The conventional voice recognition apparatus recognizes each word among words spoken by a user and indicates the result, which includes the recognized word, on the display. It is necessary to correct the result since the recognition rate of such a speech recognition apparatus is not always perfect.
FIG. 4
is a block diagram showing a construction of the conventional speech recognition apparatus, which comprises a voice inputting device
11
, a speech recognition device
12
, a recognition result holder (storage device)
13
, and a display
14
. The speech recognition device
12
automatically recognizes spoken words input from the voice inputting device
11
, and the recognition result is stored in the recognition result holder
13
and indicated on the display
14
.
The conventional speech recognition apparatus further includes an instruction inputting device
17
for inputting user instructions for corrections, and a corrector
15
for correcting the recognition result stored in the recognition result storage device
13
.
In a correction process, the display
14
indicates the recognition result, which includes the recognized word and is supplied from the speech recognition device
12
, and displays one or more candidates whose pronunciations are similar if the result includes a mis-recognized portion. The user appropriately selects one of the displayed candidates through the inputting device
17
to correct the result, so that the corrector
15
corrects the result stored in the recognition result holder
13
.
The conventional speech recognition apparatus and the method for correcting the recognition result are disclosed in Japanese Patent Application, First Publication Nos. Sho 61-177573 and Hei 1-197797.
The conventional apparatus which performs continuous word recognition, allowing a user to pronounce word strings not discretely but fluently, cannot determine breakpoints in word strings because the words are continuously spoken, and often mis-recognizes a single word as two or more words.
While in the conventional correction method some word candidates are displayed according to a user instruction, the user can not correct two or more words to a single word but must correct the single word to another single word. Therefore, the user must delete the mis-recognized portion and must enter a new word, and it is therefore difficult to efficiently input words via the voice, for example, in Japanese.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a word processor which can efficiently input words by way of voice, for example, in Japanese, even when a single word is misidentified as two or more words.
In order to accomplish the above object, the word processor of the present invention comprises: a voice inputting device for inputting a spoken word and converting the spoken word into voice data; a voice storage device for storing the voice data; a speech recognition device for recognizing a word in the voice data output from the speech inputting device or the voice data stored by the voice storage device; a display for displaying a result obtained by the speech recognition device; an instruction inputting device for inputting an instruction to select a portion in the result; and a correction device for correcting the portion in the result according to the instruction from the instruction inputting device.
In another aspect of the present invention, the word processor of the present invention further comprises a result storage device for storing the result obtained by the speech recognition device. The speech recognition device recognizes a word in the voice data from the voice inputting device in response to an input of the voice data from the voice inputting device, and recognizes a word in the voice data stored in the voice storage device in response to an input of the instruction to select a portion to be corrected. The display indicates one or more results when the speech recognition device recognizes a word in the voice data stored in the voice storage device.
The correction device corrects the selected portion to a single word. The correction device selects one of the results according to an instruction through the instruction inputting device. The instruction inputting device may be a mouse, a keyboard, or a voice command system.
A word processing method of the present invention comprises the steps of: converting input spoken words into voice data; storing the voice data data; displaying a result obtained by the recognition on a display; and recognizing a word in a selected portion of the stored voice data.
The word processing method further comprises the step of storing the result. The selected portion may be corrected to a single word. The word processing method further comprises the step of displaying a second result obtained by recognition for the selected portion. The word processing method further comprises the steps of: displaying a plurality of second results for the selected portion; and selecting one of the second results according to a user instruction.
A computer readable medium of the present invention containing program instructions performs the steps of: converting an input spoken words into voice data; storing the speech data; recognizing a word in the voice data; displaying a result obtained by the recognition on a display; and recognizing a word in a selected portion of the stored voice data.
The computer readable medium further performs the storing of the result. The selected portion may be corrected to a single word. The computer readable medium further performs display of a second result obtained by recognition for the selected portion. The computer readable medium further performs displaying a plurality of second results for the selected portion; and selecting one of the second results according to a user instruction.
According to the present invention, because portions specified by a user are recognized as a single word, two or more words mis-recognized in the result of continuous word recognition can be corrected to a single word. This eliminates the operations deleting and re-inputting words, thereby making inputting by way of voice easy.


REFERENCES:
patent: 54-148304 (1979-11-01), None
patent: 61-177573 (1986-08-01), None
patent: 1-197797 (1989-08-01), None
patent: 4-254896 (1992-09-01), None
IBM™ (“Voice Type™ Dictation for OS/2 User's Guide©”, Nov. 1994).*
McKenzie (“VoiceType™ Simply Speaking Gold,” product review for WindoWatch Sep. 28, 1997).*
Rye (“VoiceType™ Dictation 3.0 FAQ,” Rye Computer Corp. support site, Jun. 1999).*
IBM-UK (“ViaVoice™ Support,” IBM™ Support website, May 1999).

Affiliated with

Also associated with

Rating

Word processor via voice does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Word processor via voice, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Word processor via voice will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2439117