Data processing: speech signal processing – linguistics – language – Linguistics – Dictionary building – modification – or prioritization
Reexamination Certificate
1998-03-05
2002-02-05
Thomas, Joseph (Department: 2644)
Data processing: speech signal processing, linguistics, language
Linguistics
Dictionary building, modification, or prioritization
C707S793000
Reexamination Certificate
active
06345245
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a natural language processing system having such functions as word processing, machine translation, character recognition, abstract creation, voice recognition, voice synthesizing, etc., and more particularly, to a natural language processing system comprising a plurality of connected natural language processing systems.
Furthermore, the present invention relates to a common dictionary management system for use with a natural language processing system comprising a plurality of connected natural language processing systems.
2. Description of the Related Art
Current technology allows information to be transmitted in a variety of media including facsimile images, voice, and international electronic mail. Further, various forms of language information are used for those media. Under such circumstances, it is expected that the environment that enables information to be processed using technologies that convert information transmitted over such media to information understandable by unspecified computers, as well as the use of language processing technologies that converts and abstracts languages freely, will become increasingly important and required.
Conventionally, such technologies have been realized as specific systems for translating electronic mail transmitted from overseas from English into Japanese and as text and text converted to voice. In other words, such electronic mail is first translated using an English-Japanese translation system provided by an English-Japanese dictionary. Then, the resulting Japanese text is converted into speech using a voice synthesizing system provided by a Japanese voice synthesizing dictionary in a series of processings. In such a system, however, both the translation and voice synthesizing systems have dedicated dictionaries, and no dictionary is shared by either system, although a system that can execute the above series of processings automatically is already materialized for them. Thus, data must be registered/deleted into/from those translation and voice synthesizing dictionaries individually. Even when a Japanese word corresponding to a new English word is registered in the translation dictionary, the “pronunciation” and “accent” of the Japanese word must separately be registered in the voice synthesizing dictionary. And, as often seen in recent years, when one person uses a plurality of computers, for example, when that person uses a portable terminal in addition to an office desktop type computer in an office, the same registration operation must separately be performed for the dictionaries of both the translation systems provided in the desktop computer and in the portable terminal.
To facilitate such operations, one may copy the dictionaries, store functions in a server for use, and/or do other necessary operations. In the former example, in some natural language processings, the environment for using natural language processings can be shared among computers by copying user dictionaries. In the latter example, natural language processing functions can be stored in a server so that the same natural language processing can be used from any terminal. However, both methods can execute those natural language processings only in a closed system (e.g., either translation or voice synthesizing only), but no dictionary registration is possible over different types of systems. For example, when translation and voice synthesizing systems are used as explained above, a Japanese word corresponding to a new English word may be registered in the translation dictionary. In this case, however, a text including the new word cannot be voice-synthesized. To voice-synthesize the new word, the word must also be registered in the voice synthesizing dictionary separately.
Even when a natural language processing system includes a plurality of natural language processing systems connected to each other, such as word processing, machine translation, abstract creation, voice recognition, voice synthesizing, etc., new data must be registered in each dictionary separately. So far it has been impossible to reduce the work of such registration.
SUMMARY OF THE INVENTION
Under such circumstances, it is an object of the present invention to provide a common dictionary management system that can automatically register information changed in a specific dictionary of a natural language processing system in the dictionaries of other natural language processing systems.
It is another object of the present invention to provide a common dictionary management system that can specify the range of dictionaries in which changed information will be registered when the changed information in a specific dictionary of a natural language processing system is automatically registered in the dictionaries of other natural language processing systems.
In other words, the common dictionary management method of the present invention is a method for managing a common dictionary by collecting dictionary data of the local user dictionary belonging to each of a plurality of language processing systems and distributing dictionary data to those local user dictionaries, wherein dictionary data updated in any local user dictionary belonging to one of a plurality of the language processing systems is collected and the common dictionary is updated according to this collected dictionary data. Furthermore, the dictionary data of this updated common dictionary is distributed to the local user dictionaries of the language processing systems that satisfy preset conditions. Thus, a change generated in the dictionary information of a natural language processing system is automatically extended to the dictionaries of other natural language processing systems, and furthermore, the work of data registration into the dictionaries made for each natural language processing function conventionally can be reduced significantly by controlling the dictionary range to which the change is extended.
Furthermore, the common dictionary management method of the present invention is a method for managing a common dictionary by collecting dictionary data from local user dictionaries belonging to each of a plurality of natural language processing systems and distributing dictionary data to those local user dictionaries, wherein dictionary data updated in any local user dictionary belonging to one of a plurality of the language processing systems and the updating time information in the dictionary containing the updated dictionary data are collected, and the common dictionary storing dictionary data of the local user dictionaries of each of the language processing systems is updated according to this collected dictionary data and the updating time information of the collected dictionary data. Then the dictionary data information most recently distributed to the dictionaries of the language processing systems is compared with the most recent dictionary data information of the common dictionary. After this, the dictionary data of the common dictionary, selected according to the result of the comparison, is distributed to the dictionaries of the language processing systems that satisfy preset conditions, so that a change generated in dictionary information of a natural language processing system is automatically extended to the dictionaries of other natural language processing systems. Moreover, the work of conventionally registering data into dictionaries for each natural language processing function is reduced significantly by controlling the range of dictionaries to which the change is extended.
Furthermore, the common dictionary management system of the present invention is a system for managing a common dictionary by collecting dictionary data of the local user dictionaries belonging to each of a plurality of language processing systems and distributing dictionary data to those local user dictionaries, wherein the common dictionary management system comprises a common dictionary storing dictionary data of the dictionaries of a plurali
Kimura Kazuhiro
Sugiyama Hiroshi
Tsujimoto Shuichi
Oblon & Spivak, McClelland, Maier & Neustadt P.C.
Thomas Joseph
LandOfFree
Method and system for managing a common dictionary and... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and system for managing a common dictionary and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for managing a common dictionary and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2970002