JP2002091858A - Information providing device, information generator, information providing system connected therewith, method therefor and recording medium recorded with program therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information providing device, by which a user can acquire only an irreducibly minimum vocabulary dictionary required for speech operation. SOLUTION: In accordance with the transmission request of information from terminal equipment, an information providing part 210 converts a part of a display text character string in designated information into speech vocabulary dictionary data to be used for speech operation and adds these data to information corresponding to the transmission request. Then, the information is transmitted through communication equipment 230 to the terminal equipment. Therefore, the user can acquire only the irreducibly minimum vocabulary dictionary required for the speech operation.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice operation function and a voice input function used when accessing information such as a homepage provided by an Internet infrastructure, and more particularly to a voice vocabulary dictionary used for these functions. Providing information via the Internet, an information generating device for generating a speech vocabulary dictionary, an information providing system connecting them, a method thereof, and a recording medium on which such a program is recorded.

[0002]

2. Description of the Related Art In recent years, with the spread of the Internet infrastructure, it has become possible for general users to easily obtain and use information and services distributed by information service companies and the like. On the other hand, a man-machine interface of an information processing device such as a computer used to obtain such information is still very difficult for general users to understand and use.

[0003] Conventionally, in order to improve the man-machine interface, each computer maker or the like has been developing one that applies a voice recognition technique represented by a computer operation using voice, for example. As one of them, various man-machine interfaces using voice recognition have been developed in the information providing service field, and are disclosed in JP-A-8-6589, JP-A-8-223309 and JP-A-2000-76040. There is a disclosed invention.

According to the telephone line voice input system disclosed in Japanese Patent Application Laid-Open No. 8-6589, vocabulary grammar information used in each scene is extracted from a storage unit in accordance with the progress of a dialog between a user and the system. It is sent to the user terminal via the line network. The user terminal stores the received lexical grammar information in the lexical grammar buffer. The voice recognition unit reads the vocabulary grammar information from the vocabulary grammar buffer, recognizes the user's voice input signal input from the voice input unit, and sends the recognition result to the information service system via the telephone line network.

According to the voice input network service system disclosed in Japanese Patent Application Laid-Open No. 8-223309, the center system transmits voice input vocabulary information and display information stored in a storage unit to a terminal via a communication network. Send to device. The terminal device stores the voice input vocabulary information received via the communication network in the buffer, and outputs the reading information in the voice input vocabulary information and the reading information in the terminal voice input dictionary to the voice recognition unit. The recognition result output from the voice recognition unit is converted into an action and executed.

According to the voice input network terminal device disclosed in Japanese Patent Application Laid-Open No. 2000-76040, an operation command for enabling a voice operation downloaded from an application server is read, and a voice-input operation is performed. The recognition dictionary information in which information for converting the command into an inquiry sent to the application server is described and stored in association with the network address of each application server. Then, voice recognition of the user's voice input is performed based on the recognition dictionary information, and the information stored and managed is searched based on the recognition result, and a connection destination application server is specified.

[0007]

However, the above-mentioned prior art has two major problems. less than,
These problems will be described.

FIG. 17 is a block diagram showing a general system configuration of a telephone line voice input system and the like disclosed in the above-mentioned JP-A-8-6589. This system includes a terminal device 100 and an information providing server 200.
And the terminal device 100 accesses the information providing server 200 via the Internet connection provider 300 and the Internet. This system mainly uses HTML
(Hyper Text Markup Language) page data in data format is provided from the information providing server 200.

The information terminal 100 has a voice recognition unit 11
0, an action code storage unit 111, a vocabulary dictionary storage unit 112, a basic dictionary storage unit 113, a browser 130 provided as an application program, a page storage unit 131, an input device 140, and a display device 150.
And a microphone 160.

First, a network address (site address) character string data of the information providing server 200 is instructed / input from the input device 140. The browser 130 sends the information providing server 2 via the general public line and the Internet connection provider 300 based on the character string data.
00, and outputs a transmission request for information such as page data. When receiving the transmission request, the information providing server 200 transmits the page data, the vocabulary dictionary data corresponding to the page data, and the action code to the terminal device 100.

When the browser 130 receives the information,
The page data is stored in the page storage unit 131, and the display device 150 displays the page data. Further, the browser 130 outputs the vocabulary dictionary data and the action code to the voice recognition unit 110. The voice recognition unit 110 stores the vocabulary dictionary data and the action code output from the browser 130, respectively, in the vocabulary dictionary storage unit 1.
12 and the action code storage unit 111.

When the user inputs a voice via the microphone 160, the voice recognition unit 110 reads the vocabulary dictionary data stored in the vocabulary dictionary storage unit 112 and the basic dictionary storage unit 113.
The voice input data is recognized with reference to the basic dictionary stored in the. Then, the voice recognition unit 110 refers to the action code data stored in the action code storage unit 111 and generates an action code corresponding to the voice recognition result.

In the system shown in FIG. 17, the basic dictionary stored in the basic dictionary storage unit 113 is equivalent to the vocabulary dictionary stored in the vocabulary dictionary storage unit 112, and the basic dictionary is prepared in the terminal device 100 in advance. You can think that it is not necessary. However, when building a real system, it is necessary to prepare this basic dictionary. The reason will be described below.

FIG. 18 is a diagram for schematically explaining the processing contents of kana-kanji conversion. Kana-Kanji conversion is performed in two major steps: matching detection processing and sieving processing. Conversion candidates are mainly extracted by matching detection processing, and conversion candidates are narrowed down and prioritized by sieving processing. Although FIG. 18 illustrates the serial processing in which the sieving processing is performed after the matching detection processing is performed, the processing is generally performed in parallel to improve recognition accuracy and the like. However, since the problems are common, the description will be made using serial processing for simplicity.

For example, when the input data character string “Kagahaku Hanno” is input, the dictionary A shown in FIG. 19A is used in the matching detection processing, and a vocabulary that completely matches the input data character string in the registered vocabulary is used. Is checked. If there is no completely matching vocabulary, the input data character string is decomposed by the registered vocabulary registered in the dictionary A so that no extra character string is formed. As a result, a plurality of candidates such as “Scientific Reaction”, “Chemical Reaction”, and “Chemical Team Rabbit” are output.

By the sieving process, only those which are valid as sentences are selected from the plurality of candidates, or the candidate priority process is performed. FIG. 19 (b)
It is a figure showing a part of dictionary B used at the time of sieving processing. The dictionary B is a dictionary in which vocabularies that can be combined are stored. In the case of the input data character string, since only “chemical reaction” composed of a combination of “chemical” and “reaction” among the plurality of candidates detected by the matching detection processing is registered in the dictionary B, This indicates that only the “chemical reaction” is output by the sieving process.

In the above description, the Kana-Kanji conversion process is described as an example. However, the same process is performed in any recognition process such as speech recognition, handwriting recognition, natural language recognition, and the like. In this processing, there are two problems as follows.

One of them is that if the relationship between two vocabularies registered in the dictionary A is registered in the dictionary B, the information amount of the dictionary B is maximized (the vocabulary registered in the dictionary A). (Number) ² and the amount of information is enormous.

Another one is that the recognition processing of the vocabulary unit in the kana-kanji conversion processing is a phoneme unit which is a finer unit in the case of the speech recognition processing. This phoneme is a phoneme symbol representing a phoneme waveform as specific character data.
Generally, it is similar to a Roman character string. Therefore, the amount of information is further increased.

In the telephone line voice input system disclosed in Japanese Patent Application Laid-Open No. 8-6589, if it is assumed that the vocabulary dictionary is composed of phonemes, it is necessary to realize a system capable of inputting natural language voice, for example. , Dictionary B
Is not proportional to the number of recognized vocabulary words, but becomes enormous as described above, which causes problems such as a reduction in speed performance and an increase in traffic.

Conversely, assuming that the vocabulary dictionary is composed of Hiragana / Katakana character data, the possibility of obtaining incorrect recognition results is reduced depending on the control method of the recognition engine. However, there is a high possibility that a recognition result is not output at all, that is, no matter what you say, nothing happens. For example, as described in the embodiment of Japanese Patent Application Laid-Open No. 2000-76040, "Akihabara" and "Saiyudensha" of voice input data In the case of a system that performs processing to extract or remove the particle "no",
There is a problem that the capacity of the dictionary B becomes enormous and communication traffic increases every time different information is transmitted and received through the network. Also, in the case of a system that does not transmit and receive the part of the dictionary B, it is considered that the speech recognition accuracy cannot be improved at all by narrowing down the vocabulary, that is, the system cannot remove the particle "no".

Accordingly, the above-mentioned Japanese Patent Application Laid-Open No. Hei 8-6589
The technology disclosed in Japanese Unexamined Patent Publication (Kokai) No. H11-107421 is not used in a system for performing grammatical analysis, but rather uses voice commands such as performing predetermined control when voice input data completely matches the vocabulary level. Should be used in systems that use. Note that the basic dictionaries stored in the basic dictionary storage unit 113 shown in FIG. 17 correspond to phoneme waveforms and phoneme symbols corresponding to the dictionary A and phoneme symbols corresponding to the dictionary B, which are used for speech recognition processing. This is a dictionary in which the relationship between units is described. The vocabulary dictionary stored in the vocabulary dictionary storage unit 112 is stored in the information providing server 2.
This is a dictionary that describes a vocabulary string composed of phoneme symbols related to the information provided by 00.

[0023] Further, in the technology of the telephone line voice input system and the like disclosed in JP-A-8-6589, there is another one.
One problem is related to the development load of the information provider. The following four types of vocabulary dictionary data transmitted from the information providing server 200 can be considered.

Standard Speech Waveform Data Character Data Including Kanji Character Data Character Data Consisting Only of Hiragana / Katakana Characters Phoneme Symbol Data Based on Specific Regulations Consisting of Text Character Strings As described above, since the amount of transmission data is enormous, it is not suitable for the system shown in FIG.

In the above, in the terminal device 100, it is necessary to perform a kanji reading conversion process on at least the input data, and depending on the conversion logic, the operation may not match the purpose of the information provider. For example, "Yesterday" can be read as "Kino" or "Sakujitsu". On the information provider side,
“Yesterday” is to be assigned to another function “function”, so “yesterday” is to be read as “sakujitsu”. However, since it depends on the kanji reading conversion logic existing in the speech recognition engine of the terminal device 100, the information provider cannot respond.

In the above, there is a problem that the intonation of speech cannot be represented by a Hiragana / Katakana code sequence. For example, even if you try to provide a playful homepage, such as shifting to a secret page address only when you speak a predetermined vocabulary with a specific rhythm, it is impossible to realize it only by considering the information provider. Become.

Therefore, the data format of the vocabulary dictionary data is the best because it does not have these problems, and it is best that the phoneme symbols are created based on the rules agreed between the speech recognition engine providers. It is considered to be.

In recent years, an increasing number of individuals have opened homepages individually, not as businesses. Many people who browse the homepage established by such individuals are friends or close people of the establisher, that is, individuals.

Assuming that the data format of the vocabulary dictionary is unified to the above contents, it is considered that the definition of the phoneme symbols is difficult for general users to understand. With the recognition of such systems, lexical dictionary creation tools for converting to phoneme symbols will be commercialized. However, until the lexical dictionary creation tool is commercialized or the lexical dictionary creation tool is purchased, voice operations are possible when accessing the homepage for a company that has a professional homepage designer. When accessing this homepage, voice operations are not possible, which is inconsistent with the original purpose.

The present invention has been made to solve the above problems, and a first object of the present invention is to provide an information providing apparatus which enables a user to obtain only a minimum vocabulary dictionary necessary for voice operation. , An information generating apparatus, an information providing system connecting them, a method thereof, and a recording medium on which the program is recorded.

A second object is to record an information providing device, an information generating device, an information providing system connecting them, a method thereof, and a program thereof in which a user can access a linked file by voice operation. The purpose of the present invention is to provide a recording medium.

[0032]

According to one aspect of the present invention, an information providing apparatus receives a request for transmitting information, and a communication unit for transmitting information corresponding to the transmission request; In response to the requested information transmission, a part of the display text string in the specified information is converted into voice vocabulary dictionary data used for voice operation, added to the information corresponding to the transmission request, and transmitted to the communication means. Information providing means for outputting.

The information providing means converts a part of the display text character string in the information designated by the user into voice vocabulary dictionary data used for voice operation, attaches it to information corresponding to the transmission request, and sends it to the communication means. Since the output is performed, the user can obtain only the minimum vocabulary dictionary necessary for voice operation.

[0034] Preferably, the information providing means merges the converted speech vocabulary dictionary data with speech vocabulary dictionary data already existing in the information corresponding to the transmission request, and outputs the merged speech vocabulary dictionary data.

Therefore, it becomes possible to add only necessary speech vocabulary dictionary data. Preferably, the information providing means converts a part of the display text string related to the hyperlink description in the designated information into voice vocabulary dictionary data used for voice operation, and associates the communication text with the hyperlink description. Output to

Therefore, the user can access the linked file by voice operation.

[0037] More preferably, the information providing means converts the speech vocabulary dictionary data already existing in the information corresponding to the transmission request into speech vocabulary dictionary data of a different format and outputs the data to the communication means.

Therefore, it is possible to cope with speech vocabulary dictionary data of different formats. More preferably, the information providing device further includes a first dictionary for converting a hiragana / katakana character string into a phoneme symbol, and the information providing means refers to the first dictionary and reads a hiragana / display character string. / Convert katakana to phoneme symbol string.

More preferably, the information providing device further includes a second dictionary for converting kanji into Hiragana / Katakana character strings, and the information providing means refers to the second dictionary to convert the display text character string. Kanji is converted into a Hiragana / Katakana character string, and the Hiragana / Katakana character string is converted into a phoneme symbol string with reference to the first dictionary.

According to another aspect of the present invention, an information generating apparatus includes: an information generating unit configured to generate information including a plurality of files having display text strings; and a plurality of information generated by the information generating unit. A vocabulary dictionary generating unit for converting a part of the display text string of the file into voice vocabulary dictionary data used for voice operation and providing the voice vocabulary dictionary data to a plurality of files; Communication means for transmitting and uploading a plurality of files.

The vocabulary dictionary generating means converts a part of the display text character strings of the plurality of files generated by the information generating means into voice vocabulary dictionary data used for voice operation, and assigns it to the plurality of files. By uploading the information to the information providing device, it is possible to provide the user with information that can perform voice operations.

Preferably, the vocabulary dictionary generating means converts a part of the display text character string related to the hyperlink description in the plurality of files into voice vocabulary dictionary data used for voice operation, and associates it with the hyperlink description. Output to the communication means.

Therefore, the user can access the linked file by voice operation.

According to still another aspect of the present invention, a terminal device, an information providing device for providing information in response to a transmission request for information from the terminal device, and an information for generating information to be uploaded to the information providing device An information providing system including a generating device, the information generating device includes: an information generating unit configured to generate information including a plurality of files having a display text string; and a plurality of information generated by the information generating unit. A vocabulary dictionary generating unit for converting a part of the display text string of the file into voice vocabulary dictionary data used for voice operation and providing the voice vocabulary dictionary data to a plurality of files; First communication means for transmitting a plurality of files and uploading the files to the information providing apparatus, the information providing apparatus comprising: A second communication unit for receiving and transmitting information corresponding to the transmission request, and outputting the uploaded information to the second communication unit in response to the transmission request for the information received by the second communication unit Information providing means for performing

The information generating apparatus converts a part of the display text character strings of a plurality of files into voice vocabulary dictionary data used for voice operation, attaches the plurality of files to the plurality of files, and uploads the files to the information providing apparatus. It is possible to provide information that allows voice operation.

More preferably, the vocabulary dictionary generating means comprises:
A part of the display text string related to the hyperlink description in the plurality of files is converted into speech vocabulary dictionary data used for voice operation, and is output to the first communication means in association with the hyperlink description.

Therefore, the user can access the linked file by voice operation.

According to yet another aspect of the present invention, an information providing method includes a step of receiving an information transmission request, and a display text string in designated information in response to the received information transmission request. Is converted into voice vocabulary dictionary data used for voice operation and added to information corresponding to the transmission request, and a step of transmitting information to which the voice vocabulary dictionary data is added.

A part of the display text character string in the information specified by the user is converted into vocabulary vocabulary dictionary data used for voice operation and added to the information corresponding to the transmission request and transmitted. It is possible to obtain only the minimum vocabulary dictionary necessary for the user.

According to still another aspect of the present invention, an information generating method includes a step of generating information including a plurality of files having a display text string, and a step of generating a display text string of the plurality of generated files. The method includes a step of converting part of the data into voice vocabulary dictionary data used for voice operation and providing the data to a plurality of files, and a step of transmitting and uploading a plurality of files provided with the voice vocabulary dictionary data.

A part of the generated display text character strings of the plurality of files is converted into speech vocabulary dictionary data used for voice operation, and is added to the plurality of files. By uploading the information to the information providing device, In addition, it is possible to provide the user with information for performing voice operations.

According to still another aspect of the present invention, there is provided a computer-readable recording medium having recorded thereon a program for causing a computer to execute the information providing method,
The information providing method includes the steps of receiving an information transmission request, and, in response to the received information transmission request, converting a part of a display text character string in the specified information into voice vocabulary dictionary data used for voice operation. The method includes a step of converting and adding the information corresponding to the transmission request, and a step of transmitting the information to which the speech vocabulary dictionary data is added.

A part of the display text character string in the information specified by the user is converted into speech vocabulary dictionary data used for voice operation, and added to information corresponding to the transmission request and transmitted. It is possible to obtain only the minimum vocabulary dictionary necessary for the user.

According to still another aspect of the present invention, there is provided a computer-readable recording medium recording a program for causing a computer to execute the information generating method,
The information generating method includes a step of generating information including a plurality of files having display text strings, and converting a part of the generated display text strings of the plurality of files into voice vocabulary dictionary data used for voice operation. And a step of transmitting and uploading the plurality of files to which the speech vocabulary dictionary data has been added.

A part of the generated display text strings of a plurality of files is converted into speech vocabulary dictionary data used for voice operation and is added to the plurality of files. By uploading the information to the information providing device, In addition, it is possible to provide the user with information for performing voice operations.

[0056]

FIG. 1 is a diagram showing a data format of information provided in an information providing system according to an embodiment of the present invention. This data format is specially defined in the present embodiment, and is a data format of information provided on the Internet, such as HTML or XML.
(EXtensible Markup Language). This is different from the actual system in that the control method is detailed, for example, the vocabulary dictionary management area described later is not transmitted or excluded when transmitting information from the server. However, this does not hinder the essence of the present invention, but rather means that the present invention is not limited to a communication infrastructure such as the Internet or a network environment.

The data format shown in FIG. 1 is roughly composed of two areas: a provided information description area and a vocabulary dictionary information description area. The provided information description area is an area in which the content of provided information is described, and includes a display data area,
Action object management area. The display data area is an area in which text data to be displayed is described. The action object management area is an area in which information for managing an action object in which some kind of display change occurs due to some operation of a viewer is described. For example, an object in which controls such as buttons and links are performed. Is described.

For example, in the case of the HTML file shown in FIG. 2, "test:", "here" and "click here" are the display data, and the others are action objects. That is, the actual H
In the case of a TML file, normal display data, action objects, and other mere definition statements are continuously described, but in this embodiment, it is assumed that area management is performed for ease of explanation. I do.

The action object management area includes, for each action object, the type of the action object, the position and area of the action object on the corresponding display data, and the content of the action object. The type ATY of the action object indicates a type such as a button or a link. It may be considered that it corresponds to the “tag” of the HTML file. The POS indicating the position or area of the action object is added because it is managed separately from the display data area. It can be considered as a field indicating which text data on the display data area is display data of the corresponding action object.

The action object content ODT describes the control content of the action object and the like. In the present embodiment, the type A of the action object
Only the case where the TY is “link” is illustrated, and the URL (Un
It is assumed that a character string indicating an iform Resource Locator (name) is described. Although the number of action objects described is not shown in FIG. 1, it is assumed that the number is stored in a variable N secured in the information file.

The vocabulary dictionary information description area includes a vocabulary dictionary area and a vocabulary dictionary management area. The vocabulary dictionary area is an area in which vocabulary dictionary data for speech recognition corresponding to the information described in the provided information description area is described. This is the area to be stored. The vocabulary dictionary management area is an area in which management information for managing information such as whether a speech vocabulary dictionary is stored is described.

The vocabulary dictionary area includes an array variable JDATA storing a vocabulary data string that is a phoneme symbol string or a hiragana / katakana character string, and a variable ONO indicating which action object corresponds to the vocabulary data string. And In FIG. 1, the variable ONO is
For convenience, it is assumed that the values are stored in the order in which the action objects appear. Similar to the number of action objects described above, the number of vocabulary data stored in FIG. 1 is not shown, but it is assumed that the vocabulary data is stored in a variable M secured in the information file.

In the vocabulary dictionary management area, whether or not there is one or more vocabulary data in the information file, and if the vocabulary data exists, a vocabulary dictionary is automatically generated when the vocabulary data is requested to be transmitted. / Variable JFG indicating whether or not to merge, and variable JTY indicating whether currently registered vocabulary data is a phoneme symbol string or a hiragana / katakana character string.

FIG. 3 is a block diagram showing a schematic configuration of the information providing system according to the embodiment of the present invention. This information providing system includes a terminal device 1, an information providing device (hereinafter, also referred to as an information providing server) 2, an Internet connection provider 3, and an information generating device 4. Access the information providing server 2 or the information generating device 4 via the Internet.

Further, the terminal device 1 includes a voice recognition unit 10
Action code storage unit 11 and vocabulary dictionary storage unit 12
A basic dictionary storage unit 13, a browser 30 provided as an application program, and a page storage unit 31
, An input device 40, a display device 50, and a microphone 60. The input device 40 includes a keyboard, a mouse, and the like.

As will be described later, when a terminal device requests transmission of an information file, the information file described in the data format shown in FIG. 1 is transmitted from the information providing server. Request file transmission,
The processing procedure of the terminal device 1 for browsing / displaying will be described.

FIG. 4 is a flowchart for explaining the processing procedure of the terminal device 1, mainly the browser 30, in the embodiment of the present invention. First, the browser 30 determines whether or not the user has specified a URL via the input device 40 such as a keyboard (S1). When the URL is specified by the user (S1, Yes), the browser 30 determines whether the voice recognition unit 10 is operating (S2). This determination is performed by the browser 3 implemented as a PC application software form that can be incorporated (installed) even in a terminal device that does not have a speech recognition engine (in this embodiment, the speech recognition unit 10).
The control takes into account 0.

In general, the voice recognition engine also takes a form that can be incorporated into a PC system (a software form generally called driver software as distinguished from application software). That is, this determination control is, more precisely, a control that means “whether or not the speech recognition engine is incorporated in the system” in the actual system of the PC, rather than “whether or not it is operating”, The actual code of this control is processed via a software interface defined as a PC system (OS) that mediates between application software and driver software. Since the format is specified by the contents shown in FIG. 1, the description is simplified.

If the voice recognition unit 10 is operating (S2,
Yes), the process proceeds to step S7. If the voice recognition unit 10 is not operating (S2, No), a request for transmitting the information file of the specified URL is output without requesting the transmission of the voice vocabulary dictionary (S3), and the process proceeds to step S8. Advances.

In step S 1, U
If the RL is not specified (S1, No), the browser 30 determines whether or not the voice recognition unit 10 is operating (S4). If the voice recognition unit 10 is not operating (S4,
No), the process returns to step S1, and the subsequent processes are repeated. If the voice recognition unit 10 is operating (S4, Ye
s) It is determined whether a vocabulary dictionary corresponding to the currently displayed display information has already been received (S5). If the vocabulary dictionary of the display information has not been received yet (S5, No),
Returning to step S1, the subsequent processing is repeated.

In step S5, if the vocabulary dictionary of the display information has already been received (S5, Yes), it is determined whether or not the URL specified by the voice by the user has been output from the voice recognition unit 10 (S6). ). If the URL has not been output from the voice recognition unit 10 (S6, No), the process returns to step S1 to repeat the subsequent processes. When a URL is output from the voice recognition unit 10 (S6, Ye
s) or when it is determined in step S2 that the voice recognition unit 10 is operating (S2, Yes),
The browser 30 transmits a request for transmitting the information file of the specified URL to the information providing server 2 via the general public line and the Internet connection provider 3.

In step S8, the browser 30
It is determined whether an RL information file has been received. UR
If the L information file has not been received (S8, N
o), the browser 30 waits until receiving the information file. When the browser 30 receives the URL information file (S8, Yes), it is determined whether or not a vocabulary dictionary is added to the information file. If no vocabulary dictionary is added to the information file (S9,
No), the process proceeds to step S11. If a vocabulary dictionary is added to the information file (S9, Yes),
The received vocabulary dictionary is output to the voice recognition unit 10 (S1
0). Then, the browser 30 causes the display device 50 to display the received information (S11), returns to step S1, and repeats the subsequent processing.

FIG. 5 is a flowchart for explaining the processing procedure of the voice recognition unit 10. This processing procedure corresponds to the processing of the browser 30 shown in FIG. First, the voice recognition unit 10 determines whether the user has input voice through the microphone 60 (S50). If no voice has been input (S50, No), it is determined whether or not a voice vocabulary dictionary has been output from the browser 30 (S51).
If a speech vocabulary dictionary is output from the browser 30 (S
51, Yes), the contents of the speech vocabulary dictionary (JDATA
[1] to JDATA [M]) and action codes (ONO [1] to ONO [M]) are stored in the vocabulary dictionary storage unit 12.
And registered in the action code storage unit 11 (S5
2) Return to step S1 and repeat the subsequent processing. If the vocabulary dictionary has not been output from the browser 30 (S51, No), the process returns to step S1 and repeats the process.

In step S50, if the user has input a voice through the microphone 60 (S50, Ye
s), the speech recognition unit 10 uses the vocabulary dictionary information description area JDA
The voice recognition is performed by referring to the TA, and the action code ONO corresponding to the matching voice vocabulary is extracted (S5).
3). Then, the type of the action code is “link” (ATY of the ONOth action object).
Is determined (S54).

If the type of the action code is “link” (S 54, Yes), the voice recognition unit 10 outputs the ODT value (URL data), which is the action object content corresponding to the voice recognition result ONO, to the browser 30. (S55). If the type of action code is other than "link" (S54, No), action control is performed according to the ODT value of the action object corresponding to the speech vocabulary of the speech recognition result (S56).

FIG. 6 is a block diagram showing a schematic configuration of the information providing server 2 according to the embodiment of the present invention. The information providing server 2 includes an information providing unit 210, a kanji → hirakana conversion dictionary storage unit 211, a hiragana → phoneme conversion dictionary storage unit 212, a provided information storage device 220, a communication device 230, and an input device 240. , Display device 250, and CD
And a CD-ROM drive 260 in which a ROM (Compact Disc-Read Only Memory) 270 is mounted. Further, the information providing unit 210 includes a vocabulary dictionary generating unit 215, and performs overall control of the information providing server 2. The information providing unit 210 is realized by a program (hereinafter, referred to as an information providing program). This information providing program is supplied by the CD-ROM 270 and is installed on a hard disk (not shown).

The provided information storage device 220 stores information to be provided in response to a request from the terminal device 1, and is constituted by a large-capacity disk such as a hard disk. The communication device 230 receives an information provision command from the terminal device 1 via a dedicated line or a general public line, or transmits an information file to the terminal device 1.

The kanji → hirakana conversion dictionary storage section 211 stores a dictionary used when converting kanji to hiragana / katakana among text characters included in the provided information file. Also, the hiragana / phoneme conversion dictionary storage unit 212 converts hiragana / katakana among text characters included in the provided information file into phonemes, or converts hiragana / katakana converted from kanji into phonemes. The dictionary used at that time is stored.

The input device 240 includes a keyboard, a mouse, and the like. Input device 240 and display device 2
Reference numeral 50 is used for monitoring an information user or for maintenance of an information file.

FIG. 7 is a flowchart for explaining the processing procedure of the information providing unit 210 shown in FIG. First,
The information providing unit 210 determines whether an information file transmission request has been received from the terminal device 1 via the communication device 230 (S100). If an information file transmission request has not been received (S100, No), the process waits until an information file transmission request is received. Further, if the information file transmission request is received (S100, Yes), the information providing unit 2
10 determines whether or not to transmit the information file by adding a speech vocabulary dictionary when transmitting the information file (S10).
1).

When no speech vocabulary dictionary is added (S1
01, No), the contents of the information file corresponding to the URL for which transmission was requested, that is, the variable DTEXT shown in FIG.
POS [1] to POS [N] and ODT [1] to OD
T [N] is transmitted to the requesting terminal device 1 (S10
2). Then, the process returns to the entry and repeats the processing from step S100.

When a speech vocabulary dictionary is added (S101, Yes), it is determined whether or not the variable JFG value of the information file corresponding to the URL requested to be transmitted is "1" (S103). . J of the URL that requested transmission
If the FG value is not “1” (S103, No), if there is no vocabulary dictionary in the information file (JFG = 0),
Or, a vocabulary dictionary already exists, but it is specified that the contents in the information file are automatically analyzed to generate a voice vocabulary dictionary corresponding to the action object, and merged with the already existing vocabulary dictionary and transmitted. Case (JFG
= -1), the process proceeds to the entry and the processing described later is performed.

If the JFG value of the URL requested to be transmitted is "1" (S103, Yes), it is determined whether or not the variable JTY of the information file is "0" (S10).
4). If the JTY value is not “0” (S104, N
o), vocabulary dictionaries JDATA [1] to J in the information file
Since the data format of DATA [M] is a phoneme symbol string,
A speech vocabulary dictionary is added to the contents of the information file, and the variables DTEXT, POS [1] to POS [N], O shown in FIG.
DT [1] to ODT [N], JDATA [1] to JDA
TA [M] and ONO [1] to ONO [M] are transmitted to the terminal device 1 that has requested (S105). Then, the process returns to the entry and repeats the processing from step S100.

When the variable JTY of the information file is "0"
(S104, Yes), the vocabulary dictionary JDATA
Since the data format of [1] to JDATA [M] is Hiragana / Katakana, the process proceeds to the entry to convert Hiragana / Katakana into a phoneme symbol string, and performs processing described later.

As described above, two types of data formats of the vocabulary dictionary, "phoneme symbols" and "hirakana", are assumed. The reason is that the definition of the phoneme symbol data format depends on the speech recognition engine. Because. That is, assuming a service in which an information provider adds a speech vocabulary dictionary to information provided via the Internet, when this service is first commercialized, the speech recognition engine in the terminal device uses the speech recognition engine. It is necessary to access an information providing server that depends on the recognition engine. "

For example, when using a speech recognition engine manufactured by Sharp, it means that it is necessary to access information provided by an information providing server prepared by Sharp. This means that voice operations cannot be performed unless information is accessed via the server. When such a service is provided by one company, it is considered that there is not much confusion of users. However, as such services are recognized by the public, each company will develop its own speech recognition engine, and there will be phoneme symbols with different rules for each speech recognition engine. As the situation progresses further,
The standardization of phoneme symbols will be unified among companies.

In this embodiment, two types, phoneme symbols and hiragana, have been described. However, these data formats are not particularly meaningful, and a plurality of data formats may exist as information files. This is taken into account. There can be various types of information files,
Since it is desirable to be unified on a communication medium, in the present embodiment, phoneme symbols are transmitted / received as a unified format. The information file on the storage medium is stored as a phoneme symbol string and a hiragana character string. But,
The present invention is not limited to these, and may be, for example, a character string including a kanji code, or a “specified Ver1 phoneme symbol” or a “specified Ver2 phoneme symbol”.

It should be noted that the automatic generation control of the speech vocabulary dictionary need not be performed by the server that manages / stores the information file requested to be transmitted. In the present embodiment, for the sake of convenience of explanation, automatic generation control of a speech vocabulary dictionary is performed in an information providing server which manages / records an information file. By outputting the information file to be provided and the request source of the information, the other server may provide the information file to which the speech vocabulary dictionary is added to the information requester. Further, the information file requested by the terminal device 1 may be temporarily received, and the terminal device 1 may transmit the information file to the dedicated server, thereby receiving the information file to which the speech vocabulary dictionary is added from the dedicated server. good. In the actual form of service, since the person who sends information is different from the person who manages the system that provides the information, it is necessary to generate and transmit a speech vocabulary dictionary using the dedicated server described above. It is considered general.

FIG. 8 is a flowchart for explaining the processing (entry processing) when the variable JFG of the information file requested to be transmitted is not "1" in step S103 of FIG. First, the local variable i
Is initialized by substituting "1" into (S200). And
The information providing unit 210 transfers the contents of the information file requested to be transmitted to a not-shown transmission memory (S20).
1). Although the process of changing the variable value of the information file is performed after step S201, this does not change the content of the information file stored in the provided information storage device 220, but the information file transferred to the transmission memory. Means to change the content of

Next, the information providing unit 210 determines whether or not the JFG value is "0" (S202). When the JFG value is "0" (S202, Yes), that is, when the speech vocabulary dictionary is not added, "1" is substituted for the variable M to initialize (S204). When the JFG value is not “0” (S202, No), that is, when a speech vocabulary dictionary is added, the variable M is incremented by 1 (S203).

Next, the vocabulary dictionary generation unit 215 converts a text data string (Kanji) indicated by the variable POS [i] in which the text data position / area corresponding to the action object is stored into a Hiragana / Katakana character string. To CS1 (S205), and further converts the Hiragana / Katakana character string to a phoneme symbol string to CS2 (S20).
6). The vocabulary dictionary generation unit 215 uses the character string CS2 converted into the phoneme symbol string as vocabulary dictionary data as a variable JDATA.
[M], and the current value of the variable i is stored as an action object number in the variable ONO [M] (S20).
7).

Next, the information providing unit 210 determines whether or not the value of the variable i matches the number N of action objects (S208). If the value of the variable i does not match the value of the number N of action objects (S208,
No), since there is an action object that has not been processed yet, the variable i and the variable M are incremented (S209), and the processing after step S205 is repeated.

If the value of the variable i matches the value of the number N of action objects (S208, Ye
s), the generated speech vocabulary dictionary data is added to the information file, and the variables DTEXT, POS [1] shown in FIG.
~ POS [N], ODT [1] ~ ODT [N], JDA
TA [1] to JDATA [M] and ONO [1] to O
NO [M] is transmitted to the terminal device 1 that has made the request (S2
10). Then, returning to the entry, step S100
Repeat the subsequent processing.

FIG. 9 is a diagram showing an example of generation of speech vocabulary dictionary data. The display shown in FIG. 9B is performed by the HTML file shown in FIG. 9A. When the underlined "return to top page" is clicked with the mouse, a speech vocabulary dictionary such as "sentoupe-jinimodoru" is created. That is, if the user does not input the same voice as this sentence, the system will not recognize the voice.

As described above, the display text data corresponding to the action object is not always a word, but may be a sentence. In general, it should be a word, not a sentence, depending on the information provided. Therefore, step S205 in FIG.
, A word division process may be inserted before step S205, and the conversion processes of steps S205 and S206 may be performed only on the first word character string among the divided words. In addition,
Since it is only necessary to extract the first word by word division processing, it is sufficient to implement only a noun type word dictionary.

FIG. 10 is a flowchart for explaining the processing (entry processing) when the variable JTY of the information file requested to be transmitted is “0” in step S104 of FIG. First, the information providing unit 210
Is initialized by assigning “1” to a local variable i (S
300). Then, the information providing unit 210 transfers the contents of the information file requested to be transmitted to a not-shown transmission memory (S301). Although the process of changing the variable value of the information file is performed after step S301, the content of the information file stored in the provided information storage device 220 is not changed, and the information file transferred to the transmission memory is not changed. Means to change the content of

Next, the vocabulary dictionary generation unit 215 converts the content JDATA [i] (Hiragana / Katakana) of the current speech vocabulary dictionary into a phoneme symbol string to obtain CS2 (S30).
2). The vocabulary dictionary generation unit 215 uses the character string CS2 converted into the phoneme symbol string as vocabulary dictionary data as a variable JDATA.
It is stored in [i] (S303).

Next, the information providing unit 210 determines whether or not the value of the variable i matches the number M of speech vocabulary data (S304). If the value of the variable i does not match the value of the number M of speech vocabulary data (S304, No), the variable i is incremented (S305) because there is speech vocabulary data that has not been processed yet, and step S302 is performed. Repeat the subsequent processing.

If the value of the variable i matches the value of the number M of the voice vocabulary data (S304, Yes), the generated voice vocabulary dictionary data is added to the information file and shown in FIG. Variable DTEXT, POS [1] to POS
[N], ODT [1] to ODT [N], JDATA
[1] to JDATA [M] and ONO [1] to ONO
[M] is transmitted to the terminal device 1 that has made the request (S30)
6). Then, the process returns to the entry and repeats the processing from step S100.

FIG. 11 is a block diagram showing a schematic configuration of the information generating device 4 according to the embodiment of the present invention. FIG.
The only difference is that the information providing unit 210 is replaced by the information generating unit 211 as compared with the information providing server 2 in the present embodiment shown in FIG. Therefore, detailed description of the same configurations and functions will not be repeated.

The provided information is stored in the provided information storage device 220, but the information cannot be accessed from the external terminal device 1 or the like. Further, the information stored in the provided information storage device 220 may be incomplete. When the information to be provided is completed, the information generating unit 211 transmits the information to the information providing server 2 via the communication device 230.
Upload to. The information generation unit 211 is realized by a program (hereinafter, referred to as an information generation program). This information generating program is supplied by the CD-ROM 270 and is installed on a hard disk (not shown).

FIG. 12 is a flowchart for explaining the processing procedure of the information generating device 4 shown in FIG. This flowchart is similar to the one showing roughly the procedure for creating / establishing a homepage, but additionally includes a step S24 for generating a vocabulary dictionary. First, generation and preparation of a page material of a homepage are performed (S21). The basis of the page material is document data (text data) and image data. As the image data, illustrations, photo data, animations, moving images, and the like are used. In this way, the page material of the home page is prepared as digital data that can be handled by a computer.

Software functions for generating these digital data include an editor function for generating various data, an import function for inputting data from a digital data input device such as a digital camera, and the like.
Other functions include the conversion of BMP (Basic Multilingual Plane) format files to GIF (Graphics Interchange Format) format and JPEG (Joint Photographic Experts Group) format to reduce the amount of data, and image files within the Web color range. There is also an image conversion function or the like for converting and adjusting image data within a range of a specific number of colors and a specific color code to reduce a difference between devices that browse a homepage.

In recent years, magazine supplements, software packages, and photographs, illustrations, sounds, movies, CGIs (Computer Graphs) utilizing Internet infrastructure have been developed.
Materials such as icsInterface) are also provided, and individual users can also create their own homepages using these materials.

Next, page data is generated (S2).
2). One page data is composed of a plurality of materials, and is generated by, for example, decorating illustrations and graphic data around text, or pasting image data to document data.

As software functions for generating the page data, there are a simple editing function, a layout function for determining a character / image position, a link function for pasting or embedding a material such as image data, and the like. The data format of the homepage is an HTML file, and one page data is constituted by one HTML file. In the contents of the HTML file, that is, in the page data, a used material, a display position, and the like are described by character information. When the material is image data, a file name of the image data is described by a character string. .

Next, a link between pages is set (S
23). Generally, the home page is hierarchical,
As a top page, there is a table of contents page showing the contents of the entire home page, and the top page is a parent and several pages are linked. Step S23 is a step of editing / generating the configuration of the entire home page. However, if other page data is considered to be a material, the processing content is almost the same as that of step S22.

It should be noted, however, that the material is not limited to the material created by the individual user himself, for example, the material existing in the personal computer in a file format. In other words, a link can be provided to another person's homepage.

Next, a speech vocabulary dictionary is generated (S2).
4). Details of this processing will be described later. Finally, the created homepage is registered / released (S25). This is because in the processing of steps S21 to S24, the homepage of the individual user is merely created, and a third party cannot access the homepage as it is. Generally, all the files that make up the homepage created in the processing of steps S21 to S24 are
The homepage is registered / published by uploading it to the server of the contracted provider.

FIG. 13 is a flowchart for explaining the processing procedure of step S24 in FIG. 12 in more detail. First, the information generation unit 211 causes the display device 250 to display a message as to whether or not to perform a vocabulary dictionary automatic generation process and a message prompting the type of the vocabulary dictionary to be generated. Then, when there is an input from the user through the input device 240 (S400), the information generation unit 211 determines whether or not the JFG value is “0” (S401).

When the JFG value is "0" (S401, Ye
s) That is, when the vocabulary dictionary is not automatically generated, the variable JFG and the variable JTY of the vocabulary dictionary management area are added to all the files constituting the provided information,
Update registration is performed with the same file name in the information providing storage device 220 (S402). At this time, the JFG value is “0”,
The JTY value is arbitrary. Then, the information generation unit 211 causes the display device 250 to display a message as to whether or not to create a vocabulary dictionary, and prompts the user to input.

If the user does not select his / her own vocabulary dictionary (S403, No), step S25 in FIG.
The process proceeds to. When the user selects his / her own vocabulary dictionary (S403, Yes), an editor for self-creating the vocabulary dictionary is activated to create the vocabulary dictionary. Note that this processing is not directly related to the present invention, and thus will not be described in detail.

In step S401, when the JFG value is not "0" (S401, No), that is, when the vocabulary dictionary is automatically generated, the local array variable tfil
The file names of all the files constituting the provided information are stored in e [] (S404). Note that the local variable FC stores a value obtained by subtracting 1 from the number of all files constituting the provision information.

Next, the local variable k is cleared to "0" (S405), and it is determined whether the local variable k is larger than the value of FC (S406). Local variable k is F
If the value is larger than the value of C (S406, Yes), the process proceeds to step S403 to perform the subsequent processing. If the local variable k is equal to or smaller than the value of FC (S406, N
o), proceed to the entry and perform the processing described later.

FIG. 14 is a flowchart for explaining the processing procedure of automatic vocabulary dictionary generation control. First, the vocabulary dictionary generation unit 215 generates a vocabulary dictionary file,
That is, the file contents of the file indicated by the tfile [k] value are read (S410). Then, “1” is substituted for the local variable i to perform initialization (S411).

Next, the vocabulary dictionary generation section 215 determines whether or not the value of the variable i is larger than the number N of action objects (S412). If the value of the variable i is equal to or less than the number N of action objects (S412, No),
The text data string (Kanji) indicated by the variable POS [i] in which the text data position / area corresponding to the action object is stored is converted into a Hiragana / Katakana character string to obtain CS1 (S413).

Next, it is determined whether or not the variable JTY is "0" (S414). If the variable JTY is not “0” (S414, No), that is, if the type of the vocabulary dictionary data specified by the user is a phoneme symbol, the vocabulary dictionary generation unit 215 converts the Hiragana / Katakana character string into a phoneme symbol string. To CS2 (S415). Then, the character string CS2 converted into the phoneme symbol string is stored in the variable JDATA [i] as vocabulary dictionary data, and the value of the current variable i is set as the action object number and the variable ON is set.
It is stored in O [i] (S416). Then, the variable i is incremented (S417), and the process returns to step S412 to repeat the subsequent processing. That is, a vocabulary dictionary for the next action object is generated.

In step S414, if the variable JTY is “0” (S414, Yes), that is, if the type of the vocabulary dictionary data specified by the user is Hiragana / Katakana, the vocabulary dictionary generation unit 215 The character string CS1 converted into Hiragana / Katakana is stored as lexical dictionary data in the variable JDATA [i], and the value of the current variable i is stored in the variable O as the action object number.
NO [i] is stored (S418). Then, the process proceeds to step S417.

In step S412, if the value of the variable i is larger than the number N of action objects (S
412, Yes), substituting (i-1) for the variable M (S
419). Then, the vocabulary dictionary generation unit 215 adds the variable JDATA shown in FIG. 1 to the contents of the file tfile [k].
[1] to JDATA [M], ONO [1] to ONO
[M], JFG and JTY are added and updated and registered in the provided information storage device 220 with the same file name (S42).
0). Then, the value of the variable k is incremented (S4
21), the process proceeds to the loop entry Lp shown in FIG.

FIG. 15 is a diagram for explaining the transmission path of the vocabulary dictionary in the information providing system according to the present embodiment. FIG. 15A shows a case where an information file to which a vocabulary dictionary is added is transmitted from the information providing server 2. When a request for transmitting an information file is received from the terminal device 1, the information providing server 2 outputs the information file to the information generating device 4 and outputs a request for generating a vocabulary dictionary. The information generation device 4 generates a vocabulary dictionary in response to the vocabulary dictionary generation request, and transmits an information file to which the vocabulary dictionary has been added to the information providing server 2. And the information providing server 2
The information file to which the vocabulary dictionary is added is transmitted to the terminal device 1.

FIG. 15B shows a case where an information file to which a vocabulary dictionary has been added is transmitted from the information generating device 4. When there is a request to transmit an information file from the terminal device 1, the information providing server 2 outputs the information file and the ID of the terminal device that has made the transmission request to the information generation device 4, and outputs a vocabulary dictionary generation request. The information generation device 4 generates a vocabulary dictionary in response to the vocabulary dictionary generation request,
The information file provided with the vocabulary dictionary is directly transmitted to the terminal device 1.

FIG. 15C shows a case where an information file is directly transmitted from the terminal device 1 to the information generating device 4 and an information file to which a vocabulary dictionary is added is transmitted from the information generating device 4 accordingly. ing. When a request for transmitting an information file is received from the terminal device 1, the information providing server 2 transmits the information file to the terminal device 1. Terminal device 1
Outputs an information file and a transmission request to the information generation device 4, and outputs a vocabulary dictionary generation request. The information generation device 4 generates a vocabulary dictionary in response to the vocabulary dictionary generation request, and directly transmits the information file to which the vocabulary dictionary has been added to the terminal device 1.

In the three types of information providing systems shown in FIGS. 15A to 15C, the control is slightly different depending on whether or not the vocabulary dictionary is generated for a fee. That is, only in the case of the mode shown in FIG. 15A, the information generating device 4 is not notified of the ID of the terminal device 1 requesting information provision. Therefore, in order to charge the generation of the vocabulary dictionary, the terminal device 1 needs to determine whether or not the user is a contractor.
A configuration for receiving the ID notification from the information providing server 2 or receiving the ID directly from the terminal device 1 is required. In addition, it is necessary to provide a contractor database in order to charge the service, and to determine whether or not the contractor.

FIG. 16 is a flowchart for explaining a processing procedure when the information providing service is charged. This processing is added to the information generating device 4. First, it is determined whether a service has been requested (S500). If the service has not been requested (S500, No), the process waits until the service is requested. If a service is requested (S500, Yes), it is determined whether the service requester is a contractor (S500).
501).

If the service requester is not a contractor (S
501, No), the information file output from the information providing server 2 or the terminal device 1 is transmitted to the information providing server 2 or the terminal device 1 as it is (S502). Then, the process returns to step S500 to repeat the subsequent processing.

If the service requester is a contractor (S501, Yes), the vocabulary dictionary generation processing described above is performed (S503), and the information file with the vocabulary dictionary is transferred to the information providing server 2 or the terminal device 1. (S5)
04). Then, the process returns to step S500 to repeat the subsequent processing.

In the present embodiment described above,
The explanation was mainly on the case of voice recognition of the link designation part. However, it is necessary to create a speech vocabulary dictionary of the text part other than the link designation part and perform various actions by voice operation by associating it with various action objects. Is also possible.

As described above, according to the information providing system in the present embodiment, the information providing server 2 creates a vocabulary dictionary corresponding to the display text transmitted from the terminal device 1 and transmits it to the terminal device 1 As a result, the user can obtain only the minimum vocabulary dictionary necessary for voice operation, and the data amount of the vocabulary dictionary stored in the terminal device 1 can be reduced. In addition, the information provider does not need to create a vocabulary dictionary to be transmitted, so that the burden on the information provider can be reduced.

Further, the information providing server 2 converts the display text transmitted from the terminal device 1 into phoneme symbols and transmits the phoneme symbol to the terminal device 1. It has become possible to reduce the amount of data.

Further, since the information generating device 4 converts a part of the text character string present in each file of the homepage into a speech vocabulary dictionary and gives it, the viewer can perform a voice operation. It became. In particular, by creating a voice vocabulary dictionary in association with a URL description, it has become possible to easily access a linked file by voice operation.

The embodiments disclosed this time are to be considered in all respects as illustrative and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

[Brief description of the drawings]

FIG. 1 is a diagram showing a data format of information provided in an information providing system according to an embodiment of the present invention.

FIG. 2 is a diagram illustrating an example of an HTML file.

FIG. 3 is a block diagram illustrating a schematic configuration of an information providing system according to the embodiment of the present invention.

FIG. 4 is a flowchart for explaining a processing procedure of the terminal device 1 according to the embodiment of the present invention.

FIG. 5 is a flowchart for explaining a processing procedure of the voice recognition unit 10;

FIG. 6 is a block diagram illustrating a schematic configuration of an information providing server 2 according to the embodiment of the present invention.

7 is a flowchart illustrating a processing procedure of an information providing unit 210. FIG.

FIG. 8 is a flowchart for explaining a process when the variable JFG of the information file requested to be transmitted is not “1” in step S103 of FIG. 7;

FIG. 9 is a diagram showing an example of generation of speech vocabulary dictionary data.

FIG. 10 is a flowchart illustrating a process when the variable JTY of the information file requested to be transmitted is “0” in step S104 of FIG. 7;

FIG. 11 is a block diagram illustrating a schematic configuration of an information generation device 4 according to the embodiment of the present invention.

FIG. 12 is a flowchart for explaining a processing procedure of the information generating device 4 according to the embodiment of the present invention.

FIG. 13 is a flowchart for explaining the processing procedure of step S24 in FIG. 12 in further detail;

FIG. 14 is a flowchart illustrating a processing procedure of vocabulary dictionary automatic generation control.

FIG. 15 is a diagram illustrating a transmission path of a vocabulary dictionary in the information providing system according to the embodiment of the present invention.

FIG. 16 is a flowchart illustrating a processing procedure when the information providing service is charged.

FIG. 17 is a block diagram showing a general system configuration in a conventional telephone line voice input system or the like.

FIG. 18 is a diagram for schematically explaining processing contents of kana-kanji conversion.

FIG. 19 is a diagram showing an example of a dictionary used in kana-kanji conversion processing.

[Explanation of symbols]

1 terminal device, 2 information providing device, 3 Internet connection provider, 4 information generating device, 10 voice recognition unit,
11 action code storage unit, 12 vocabulary dictionary storage unit, 13 basic dictionary storage unit, 30 browser, 31 page storage unit, 40, 240 input device, 50, 250
Display device, 60 microphone, 210 information providing unit, 211
Kanji → Hiragana conversion dictionary storage unit, 212 Hiragana →
Phoneme conversion dictionary storage unit, 214 information generation unit, 215 vocabulary dictionary generation unit, 220 information provision storage unit, 230 communication device, 260 CD-ROM drive, 270 CD-
ROM.

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI theme coat ゛ (reference) // G06F 17/30 110 G10L 3/00 571A

Claims (14)

[Claims] A communication unit for receiving a transmission request for information and transmitting information corresponding to the transmission request, wherein the communication unit receives the information transmission request and transmits the information corresponding to the transmission request. An information providing apparatus comprising: an information providing unit for converting a part of a display text character string into voice vocabulary dictionary data used for voice operation, adding the converted data to information corresponding to the transmission request, and outputting the information to the communication unit. 2. The information according to claim 1, wherein the information providing unit merges the converted speech vocabulary dictionary data with speech vocabulary dictionary data already existing in information corresponding to the transmission request, and outputs the merged speech vocabulary dictionary data. Providing device. 3. The information providing means converts a part of a display text string related to a hyperlink description in the designated information into voice vocabulary dictionary data used for voice operation, and converts the part into the hyperlink description. 3. The information providing apparatus according to claim 1, wherein the information is output to the communication unit in association with the information. 4. The information providing means converts the speech vocabulary dictionary data already existing in the information corresponding to the transmission request into speech vocabulary dictionary data of a different format and outputs the speech vocabulary dictionary data to the communication means. 4. The information providing device according to any one of claims 3 to 3. 5. The information providing device further comprises a hiragana /
A first dictionary for converting a katakana character string into a phoneme symbol, wherein the information providing unit converts the hiragana / katakana of the display text character string into a phoneme symbol string with reference to the first dictionary; The information providing device according to any one of claims 1 to 4. 6. The information providing apparatus further includes a second dictionary for converting a kanji into a hiragana / katakana character string, wherein the information providing means refers to the second dictionary and displays the display text character. 6. The information providing apparatus according to claim 5, wherein the kanji in the column is converted into a Hiragana / Katakana character string, and the Hiragana / Katakana character string is converted into a phoneme symbol string with reference to the first dictionary. 7. An information generating means for generating information including a plurality of files having a display text string, and voice operation of a part of the display text strings of the plurality of files generated by the information generating means. A vocabulary dictionary generating means for converting the vocabulary dictionary data to be used for the vocabulary dictionary and providing the vocabulary dictionary data to the plurality of files; and An information generating apparatus including: a communication unit. 8. The vocabulary dictionary generating means converts a part of a display text string associated with a hyperlink description in the plurality of files into voice vocabulary dictionary data used for voice operation, and converts the text into a hyperlink description. The information generating apparatus according to claim 7, wherein the information is output to the communication unit in association with the information. 9. An information including a terminal device, an information providing device for providing information in response to an information transmission request from the terminal device, and an information generating device for generating information to be uploaded to the information providing device. A providing system, wherein the information generating device includes: an information generating unit configured to generate information including a plurality of files having a display text string; and display text characters of the plurality of files generated by the information generating unit. A vocabulary dictionary generating means for converting a part of the sequence into voice vocabulary dictionary data used for voice operation and providing the plurality of files; and a plurality of files to which voice vocabulary dictionary data is provided by the vocabulary dictionary generating means. And first communication means for uploading the information to the information providing apparatus, and the information providing apparatus transmits information from the terminal apparatus. A second communication unit for receiving a communication request and transmitting information corresponding to the transmission request; and responding to the transmission request for the information received by the second communication unit, Information providing means for outputting to the second communication means. 10. The vocabulary dictionary generating means converts a part of a display text string associated with a hyperlink description in the plurality of files into voice vocabulary dictionary data used for voice operation, The information providing system according to claim 9, wherein the information is output to the first communication unit in association with the information. 11. A step of receiving an information transmission request, and in response to the received information transmission request, speech vocabulary dictionary data used for voice operation of a part of a display text character string in designated information. And providing the information to the information corresponding to the transmission request; and transmitting the information to which the voice vocabulary dictionary data is added. 12. Generating information including a plurality of files having display text strings, and converting a part of the generated display text strings of the plurality of files into voice vocabulary dictionary data used for voice operation. An information generation method, comprising: converting and attaching the plurality of files to the plurality of files; and transmitting and uploading the plurality of files to which the speech vocabulary dictionary data is assigned. 13. A computer-readable recording medium storing a program for causing a computer to execute the information providing method, wherein the information providing method includes a step of receiving an information transmission request; Converting a part of the display text string in the designated information into voice vocabulary dictionary data used for voice operation in response to the information transmission request, and attaching the information to information corresponding to the transmission request; Transmitting information to which vocabulary dictionary data has been added, the computer-readable recording medium. 14. A computer-readable recording medium recording a program for causing a computer to execute the information generation method, wherein the information generation method stores information including a plurality of files having a display text string. Generating, converting a part of the generated display text character strings of the plurality of files into voice vocabulary dictionary data used for voice operation, and attaching the data to the plurality of files; Transmitting and uploading the plurality of assigned files.