Data processing: speech signal processing – linguistics – language – Speech signal processing – Recognition
Reexamination Certificate
1999-09-23
2001-04-03
Dorvil, Richemond (Department: 2741)
Data processing: speech signal processing, linguistics, language
Speech signal processing
Recognition
C704S251000, C704S252000, C704S231000
Reexamination Certificate
active
06212499
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to a data retrieval apparatus, a data retrieval method and a storage medium therefor, and a communications apparatus. More particularly, the invention relates to a voice-based data retrieval apparatus having a voice recognition function, a data retrieval method applied to this apparatus, a storage medium storing a program that executes the data retrieval method, and a communications apparatus based upon this data retrieval.
2. Description of the Related Art
In order to specify the number of a party to be called from communications equipment such as a telephone or facsimile machine, the user is required to perform a manual operation, such as the pressing of dialing keys, numeric keys or registration keys provided on an equipment control panel. In the case of registration keys, the number thereof can reach several dozen to several hundred, and searching for the desired key can be a time-consuming operation. Further, such communications equipment is becoming increasingly complicated owing to the presence of a large number of operating keys, and operating a control panel that involves operating these keys is difficult for an operator who is not accustomed to operating the equipment.
Retrieval of registered information in such communications equipment and operation of the equipment have been facilitated in part by applying voice recognition and voice synthesis now being developed. Specifically, by using voice to designate the name of a party to be called, a number registered in advance is retrieved automatically to make the call. In a situation where the name of the party to be called has not been registered, the telephone number of the party may be specified by voice directly. If this is done, the communications equipment is capable of recognizing the number and of calling the number specified.
A wide variety of voice-recognition algorithms are currently available and improvements in these algorithms have led to the application of voice recognition and synthesis technology to various equipment. This is making it easier for the operator to operate such equipment. Specifically, with present-day communications equipment having a voice recognition function, the calling party pronounces a predetermined number or a registered name by voice that is recognized by the equipment. This makes it unnecessary for the operator to perform a complicated key operation using the control panel of the equipment.
However, when it is attempted to raise the voice recognition rate by accurately discriminating the voice of the operator from ambient noise and, at the same time, to make it possible to recognize the voice of an unspecified speaker, this results in the need for a complex voice-recognition algorithm and large-capacity voice dictionary as a database for voice recognition. As a consequence, a very large memory capacity is required for the equipment-control memory and database memory, thereby raising the cost of the equipment.
Though it is possible to reduce the required memory capacity by simplifying the voice recognition algorithm and scaling down the size of the voice dictionary, this is disadvantageous in that is raises the probability that the input voice will be recognized incorrectly. If the equipment misrecognizes the voice, it is necessary to input voice again or to correct the content of the input by operating keys on the control panel. Inputting the required information takes time.
SUMMARY OF THE INVENTION
Accordingly, an object of the present invention is to provide a data-retrieval apparatus, a data-retrieval method and storage medium therefor, and a communications apparatus capable of improving operability, inclusive of shortening operating time, and of holding down increases in manufacturing cost.
According to the present invention, the foregoing object is attained by providing a data-retrieval apparatus having a voice-recognition function, comprising: voice-recognition means for subjecting a syllable or phoneme at the beginning of an input voiced word to voice recognition; data-retrieval means for retrieving, from a plurality of groups of vocabulary data set in advance, applicable vocabulary data using the syllable or phoneme recognized by the voice-recognition means; and display means for displaying the vocabulary data retrieved by the data-retrieval means, thereby notifying an operator of the vocabulary data.
Further, in accordance with the present invention, there is provided a data-retrieval method applied to a data-retrieval apparatus having a voice-recognition function, comprising: a voice-recognition step of subjecting a syllable or phoneme at the beginning of an input voiced word to voice recognition; a data-retrieval step of retrieving, from a plurality of groups of vocabulary data set in advance, applicable vocabulary data using the syllable or phoneme recognized at the voice-recognition step; and a display step of displaying the vocabulary data-retrieved at the data-retrieval step, thereby notifying an operator of the vocabulary data.
Further, in accordance with the present invention, there is provided a data-retrieval apparatus comprising: input means for inputting voice; storage means for storing a plurality of data groups in units of syllables or phonemes at the beginning of the data; recognition means for recognizing a syllable or phoneme at the beginning of the voice that has been input; and means for retrieving, from the storage means, data belonging to a group corresponding to the syllable or phoneme recognized by the recognition means.
Further, in accordance with the present invention, there is provided a communications apparatus comprising: input means for inputting voice; storage means for storing a plurality of items of destination data grouped in units of syllables or phonemes; recognition means for recognizing a syllable or phoneme at the beginning of the voice that has been input; and means for retrieving, from the storage means, destination data belonging to a group corresponding to the syllable or phoneme recognized by the recognition means.
Further, in accordance with the present invention, there is provided a computer-readable storage medium storing, as a program, a data-retrieval method applied to a data-retrieval apparatus having a voice-recognition function, the data-retrieval method comprising: a voice-recognition step of subjecting a syllable or phoneme at the beginning of an input voiced word to voice recognition; a data-retrieval step of retrieving, from a plurality of groups of vocabulary data set in advance, applicable vocabulary data using the syllable or phoneme recognized at the voice-recognition step; and a display step of displaying the vocabulary data retrieved at the data retrieval step, thereby notifying an operator of the vocabulary data.
REFERENCES:
patent: 4618984 (1986-10-01), Das et al.
patent: 5832428 (1998-11-01), Chow et al.
patent: 5832429 (1998-11-01), Gammel et al.
patent: 6014624 (2000-01-01), Raman
“Dragon Dictate User's Guide”, Dragon Systems, Inc., Copyright 1986-1996, First edition 1996b.
Canon Kabushiki Kaisha
Dorvil Richemond
Fitzpatrick ,Cella, Harper & Scinto
Nolan Daniel A.
LandOfFree
Audible language recognition by successive vocabulary reduction does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Audible language recognition by successive vocabulary reduction, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Audible language recognition by successive vocabulary reduction will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2552093