Semantic recognition system

• US 5,991,719 A
• Filed: 09/11/1998
• Issued: 11/23/1999
• Est. Priority Date: 04/27/1998
• Status: Expired due to Term

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Yazaki, Masatomo, Gomi, Toshiaki, Yamamoto, Kenji, Noda, Masahide
• Primary Examiner(s): MCFADDEN, SUSAN IRIS
• Application Number:
  US09/151,291
• Time in Patent Office: 438 Days
• Field of Search: 704/250, 704/251, 704/270, 704/275
• US Class Current: 704/251
• CPC Class Codes: G06F 3/16 Sound input; Sound output speech processing G10L
• CPC Class Codes: G10L 2015/228 of application context

Abstract

A semantic recognition system of the present invention provides a user interface capable of receiving speech input to a user and an application interface that conveys an input content of the user to an application. The semantic recognition system includes a speech signal input part for receiving input speech signals, a speech recognizer for recognizing a corresponding word based on the input speech signals, a recognized word-semantic number converter including a semantic number-registered word list indicating the correspondence between a semantic number representing a meaning of a word and a registered word belonging to the semantic number, an application interface and an application handling the semantic numbers as data. The corresponding word is recognized by the speech recognizer, based on the speech signals input to the speech signal input part. The recognized word is converted to a corresponding semantic number by the recognized word-semantic number converter. The converted semantic number is notified to the application via the application interface.

Claims

1. A semantic recognition system providing a user interface capable of receiving speech input to a user and an application interface that conveys an input content of the user to an application, the semantic recognition system comprising:

a speech signal input part for receiving input speech signals;

a speech recognizer for recognizing a corresponding word based on the input speech signals;

a recognized word-semantic number converter including a semantic number-registered word list indicating a correspondence between a semantic number representing a meaning of a word and a registered word belonging to the semantic number;

an application interface; and

an application handling the semantic numbers as data,

wherein the corresponding word is recognized by the speech recognizer, based on the speech signals input to the speech signal input part,

the recognized word is converted to a corresponding semantic number by the recognized word-semantic number converter, and

the converted semantic number is notified to the application via the application interface.

2. The semantic recognition system according to claim 1, further comprising a semantic number-registered word list updater for updating a content of the semantic number-registered word list in the recognized word-semantic number converter.

3. The semantic recognition system according to claim 1, further comprising an external description file storing a word-semantic system table that systematically describes a semantic number obtained by classifying meanings systematically and coding each semantic unit and a registered word belonging to each semantic number.

4. The semantic recognition system according to claim 3, wherein

the application contains the semantic number at least in a part of process data, and

the recognized word-semantic number converter comprises a semantic number acquirer for acquiring the semantic number described in the application via the application interface, and a semantic number-registered word list generator for extracting a registered word corresponding to the acquired semantic number from the word-semantic system table in the external description file so as to generate the semantic number-registered word list.

5. The semantic recognition system according to claim 4, wherein

the recognized word-semantic number converter comprises a registered word notifier for notifying the speech recognizer of registered words listed in the generated semantic number-registered word list, and

the speech recognizer comprises a speech recognition word updater for updating a content of a speech recognition engine based on the notified registered word.

6. The semantic recognition system according to claim 5, comprising:

a use language selector for selecting a language for the user to use;

external description files each of which corresponds to each of a plurality of languages as the external description file; and

speech recognition engines each of which corresponds to each of a plurality of languages as the speech recognition engine in the speech recognizer,

wherein the semantic number-registered word list generator generates a semantic number-registered word list, based on the external description file of a language selected by the use language selector, and

the speech recognizer uses the speech recognition engine of a language selected by the use language selector,

whereby a multilingual user interface preparing the application for input instructions in a plurality of languages can be provided to the user.

7. The semantic recognition system according to claim 6 built into a client-sever system, comprising

an external description file server for providing the external description file;

an application server for providing the application;

at least one semantic recognition system client including the speech signal input part, the speech recognizer, the recognized word-semantic number converter and the application interface,

wherein the external description file server, the application server and the semantic recognition system client are connected via a computer network.

8. The semantic recognition system according to claim 5 built into a client-sever system, comprising

an external description file server for providing the external description file;

an application server for providing the application;

at least one semantic recognition system client including the speech signal input part, the speech recognizer, the recognized word-semantic number converter and the application interface, wherein the external description file server, the application server and the semantic recognition system client are connected via a computer network.

9. The semantic recognition system according to claim 3, wherein

the application contains the semantic number at least in a part of process data, and comprises the process steps of:

extracting a corresponding registered word from the word-semantic system table in the external description file, based on the semantic number described in the application, and

notifying the recognized word-semantic number converter of

the semantic number and the registered word, and

the recognized word-semantic number converter comprises a semantic number-registered word list generator for generating the semantic number-registered word list, based on the semantic number and the registered word acquired from the application.

10. The semantic recognition system according to claim 9, wherein

the recognized word-semantic number converter comprises a registered word notifier for notifying the speech recognizer of registered words listed in the generated semantic number-registered word list, and

the speech recognizer comprises a speech recognition word updater for updating a content of a speech recognition engine based on the notified registered word.

11. The semantic recognition system according to claim 10, comprising:

a use language selector for selecting a language for the user to use;

external description files each of which corresponds to each of a plurality of languages as the external description file; and

speech recognition engines each of which corresponds to each of a plurality of languages as the speech recognition engine in the speech recognizer,

wherein the semantic number-registered word list generator generates a semantic number-registered word list, based on the external description file of a language selected by the use language selector, and

the speech recognizer uses the speech recognition engine of a language selected by the use language selector,

whereby a multilingual user interface preparing the application for input instructions in a plurality of languages can be provided to the user.

12. The semantic recognition system according to claim 11 built into a client-sever system, comprising

an external description file server for providing the external description file;

an application server for providing the application;

at least one semantic recognition system client including the speech signal input part, the speech recognizer, the recognized word-semantic number converter and the application interface,

wherein the external description file server, the application server and the semantic recognition system client are connected via a computer network.

13. The semantic recognition system according to claim 10 built into a client-sever system, comprising

an external description file server for providing the external description file;

an application server for providing the application;

at least one semantic recognition system client including the speech signal input part, the speech recognizer, the recognized word-semantic number converter and the application interface,

wherein the external description file server, the application server and the semantic recognition system client are connected via a computer network.

14. A computer-readable recording medium where a semantic recognition process program is recorded, the program describing process steps for realizing a system for providing a user interface capable of receiving speech input to a user and an application interface that conveys an input content of the user to an application,

the program comprising the process steps of:

a speech signal input process step of receiving input speech signals;

a speech recognizing process step of recognizing a corresponding word based on the input speech signals;

a semantic number-registered word list generating process step of generating a semantic number-registered word list indicating the correspondence between a semantic number representing a meaning and a registered word belonging to the semantic number;

a recognized word-semantic number converting process step of converting a word recognized at the speech recognizing process step to a corresponding semantic number with reference to the generated semantic number-registered word list; and

a semantic number notifying process step of notifying the application of the converted semantic number.