JPH11134169A - Document retrieval system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable text-voice conversion suitable to a document retrieval system making good use of a network by providing a server terminal with a voice converting means. SOLUTION: When the address of information which desires to be retrieved and displayed is inputted by using a display input means 11, the input address is sent to a document retrieval means 2, a program and a button for voice reproduction which are stored in a voice load program storage means 7 are embedded in a document obtained by retrieval, and data on them are outputted to a client terminal having sent the retrieval request out. When there is not the voice data corresponding to the retrieved address in a voice storage means 3, a sentence dividing means 6 divides the text data into sentences, a voice converting means 4 converts them into voice data, and the voice storage means 3 stores them in a directory. When the mouse cursor is clicked on the voice reproduction button on the display screen of the client terminal, the reproducing program is started and corresponding voice data are downloaded from a voice conversion server 10 and reproduced as a voice.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] The present invention relates to a document search function,
The present invention relates to a document search system having a voice conversion function of converting a search document into a voice format.

[0002]

[Prior Art] Literature: Oki Electric R & D No.152 Vol.5
8 No.4 P85-88 "Development of text-to-speech conversion technology" The above-mentioned documents describe text-to-speech conversion technology that obtains a synthesized sound with natural sound quality from text data. This type of conversion technique can be classified into a conversion technique using software processing and a conversion technique using dedicated hardware, and both of them are described in the above-mentioned literature.

[0003]

However, since the conversion technology described in the above-mentioned document is basically a conversion technology for voice conversion on a single device or a computer, the conversion technology on a network. When applying to a text-to-speech conversion system that converts a retrieved document into speech on each of a plurality of computers, it is necessary to install software or hardware for text-to-speech conversion on each computer serving as a client. there were.

The present invention has been made in view of the above points, and aims to propose a text-to-speech conversion technique suitable for a document retrieval system using a network.

[0005]

According to a first aspect of the present invention, there is provided a document search system in which a plurality of client terminals and a plurality of server terminals are connected via a network. To do.

That is, the server terminal (1) retrieves the necessary information from the address on the network for which the retrieval instruction has been issued, and (2) converts the text data contained in the retrieved information into voice data. Voice conversion means for converting; (3) voice storage means for storing voice data; (4)
A terminal control means for transmitting the extracted information is provided to the client terminal which has issued the search instruction.

In a second aspect of the present invention, the following means is provided in a document search system in which a plurality of client terminals and a server terminal are connected via a network.

[0008] That is, the server terminal (1) a document retrieval means for retrieving necessary information from an address on the network instructed to retrieve, and (2) a text for storing text data contained in the retrieved information. Data storage means, and (3) voice conversion means for reading out the text data from the text data storage means and sequentially converting the text data into voice data when voice reproduction of the text data included in the extracted information is instructed; ) The client terminal which has issued the search instruction is provided with terminal control means for transmitting the extracted information.

As described above, in the first and second aspects of the present invention, since the voice conversion means is provided in the server terminal, even when text data contained in the searched information is to be reproduced as voice, Each client terminal may access the server terminal and reproduce the audio data converted by the audio conversion means.

As a result, as in the case of the conventional system,
It is not necessary to provide the voice conversion means in each of the client terminals, and the load on the client terminals can be reduced.

[0011]

BEST MODE FOR CARRYING OUT THE INVENTION

(A) Relationship Between Embodiments Hereinafter, embodiments of a document search system according to the present invention will be described by dividing them into four representative embodiments. FIG.
Is a summary of the features of the four embodiments, focusing on the differences. In any of the embodiments, the voice conversion software (or device) adopts a configuration in which only one is provided in the server, and the text-to-speech conversion is executed in the server.

The first embodiment is characterized in that audio data converted by audio conversion software (or apparatus) is stored on the server side, and a reproduction program for the stored audio data is stored on the server side. The point to store. In the first embodiment, at the time of audio reproduction, the reproduction program is downloaded from the server to the client, and the audio data in the server is read based on the reproduction program.

The feature of the second embodiment is the same as that of the first embodiment in that the converted voice data is stored in the server side in common. However, the difference is that a reproduction program for reproducing audio data is stored in each client.

A feature of the third embodiment is that the server is not stored in the form of voice data after voice conversion as in the first and second embodiments, but is stored in the form of text data divided in units of one sentence. It differs in that it is stored within. In addition, a reproduction program for reproducing the audio data obtained by converting the text data is stored in the server, and when reproducing the audio data, the reproduction program is downloaded from the server to the client, and the audio data is reproduced. Take.

The feature of the fourth embodiment is that it is stored in the server in the form of text data divided in units of one sentence.
This is the same as the third embodiment. However, the difference is that a reproduction program for reproducing audio data is stored in each client. Hereinafter, embodiments having these features will be described with reference to the drawings.

(B) First Embodiment (B-1) Configuration of System According to First Embodiment FIG. 1 is a block diagram showing a functional configuration of a document search system according to the first embodiment.

The search system according to the first embodiment has one server terminal and a plurality of terminals connected thereto (FIG. 1).
3) client terminals. In FIG. 1, the voice conversion server 10 corresponds to a server terminal, and a set of the display input unit 11 and the voice reproducing unit 12 corresponds to a client terminal.

First, the configuration of the voice conversion server 10 will be described. The voice conversion server 10 includes a network connection unit 1
It comprises a document search means 2, a voice storage means 3, a voice conversion means 4, a terminal control means 5, a sentence division means 6, and a voice load program storage means 7.

The network connection means 1 is a means for enabling communication between the voice conversion server 10 and another terminal connected via a network. In the case of FIG. 1, the network connection means 1 includes two document search means 8 and 9
Is used for communication of search instructions from the voice conversion server 10 and communication of searched documents.

The document search means 2, 8 and 9 are provided with a URL (Un
This is a means for searching for an information address on the network specified by iform Resource Locator) or the like, and taking in a document held in association with the address.

The document search means 2 is provided with a function of embedding a sound reproduction button (consisting of an icon or the like) and a program required for the reproduction in a searched document.

The voice storing means 3 is a means for storing the voice data converted by the voice converting means 4 for a certain period.

The voice converting means 4 is means for converting text data on the retrieved document into voice data. FIG.
FIG. 1 shows an example of such a voice conversion unit 4. The voice conversion means 4 is described in the above-mentioned document, and is composed of a text analysis unit 4A, a synthesis parameter generation unit 4B, a voice synthesis unit 4C, a word dictionary 4D, and segment data 4E. I have.

The terminal control means 5 is a means for individually controlling the connection between the voice conversion server 10 and each client terminal.

The sentence dividing means 6 is a means for extracting only a text portion from the searched document and dividing the text for each sentence. For example, in the case of a Japanese sentence, a period, a line feed, and the like are regarded as one sentence.

The voice load program storage means 7 is a storage means for storing a program necessary for the client terminal to download voice data from the voice conversion server 10 and control reproduction thereof.

Next, the configuration of the client terminal will be described. The client terminal includes a display input unit 11 (*) and a sound reproducing unit 12 (*) (where * = A, B, C).
It is composed of

Here, the display input means 11 (A), (B)
And (C) are means used for inputting the address of the document to be searched on the network and giving it to the speech conversion server 10, and for displaying the document searched by the speech conversion server 10. Further, the display input means 11
In (A), (B) and (C), when a sound reproduction button on the screen is operated for a document in which a reproduction program is embedded, the program contained in the document is executed to execute sound data. There is also provided a function necessary to read out.

The sound reproducing means 12 (A), (B) and (C) read out the sound data stored in the sound conversion server 10 from the text data in the document and reproduce the sound data. Means.

(B-2) Example of Initial Display Screen Next, FIG. 4 shows a basic layout configuration of a display screen displayed on the screen of each display input means 11 of the client terminal according to the first embodiment.

The largest window 13 in FIG.
Is a window of the display input means 11. At the top of the window 13, there is provided a window 14 for inputting an address on the network where information to be searched is stored. Enter the address here,
For example, a keyboard or the like connected to the client terminal is used. In FIG. 4, "http: //www.foo.c
o.jp/index.html "is entered as the address.

In the case of FIG. 4, a window 15 for displaying information as a search result based on the input address is displayed at the bottom of the window 14.

(B-3) Voice Conversion Operation FIG. 5 illustrates a voice conversion operation by the document search system according to the first embodiment.

First, it is assumed that a user using a certain client terminal has input an address of information to be searched and displayed using the display input means 11. The address input in step SP1 is sent to the document search means 2 through the terminal control means 5.

In the following step SP2, the document search means 2 operates to search for information corresponding to the address. However, if the corresponding address does not exist in the same server, the information is transmitted via the network connection means 1. It connects to the network, searches for information of the address to be searched using the other document search means 8 and 9, and receives the result.

At step SP3, the document search means 2
The voice reproduction program and buttons stored in the voice loading program storage means 7 are embedded in the document obtained as a result of the search, and the data is output to the client terminal which issued the search request via the terminal control means 5. .

In the next step SP4, the document search means 2
Confirms whether there is voice data corresponding to the searched address in the voice storage means 3. Here, if a positive result is obtained and it is confirmed that the user already has the information, the processing ends.

On the other hand, if a negative result is obtained, the process proceeds to step SP5, where the document search means 2 sends the searched information to the sentence division means 6. Here, the sentence division unit 6 extracts only the text part of the information, performs a process of dividing the extracted text data into one sentence at a time, and returns it to the document search unit 2.

When the division processing is completed, step S
Proceeding to P6, the document search means 2 sends the divided text data to the voice conversion means 4. The voice conversion means 4 converts the received text data into voice data and returns it to the document search means 2.

In step SP7, the document search means 2 passes the voice data to the voice storage means 3. The voice storage means 3
Store the received audio data in the directory. The location of the directory is uniquely determined by the searched address. The audio data is filed for each sentence, and is assigned a sequential number.

At step SP8, the voice storage means 3
Creates an audio conversion end file in the directory indicating that all the audio files have been stored, and ends a series of audio conversion operations.

(B-4) Example of Display Screen after Voice Conversion FIG. 6 shows an example of a display screen appearing on the display screen of the client terminal that has requested information retrieval after the above-described voice conversion.
FIG. 6 shows the same parts as in FIG.

The difference between FIG. 6 and FIG. 4 is that, in the case of FIG. 6, information (including text) 17 corresponding to the input address is used.
Is displayed in the window 15, and a voice reproduction button (icon) 16 for instructing to start reading a text portion of the displayed information is displayed in the upper right and left portions of the window 15. It is.

For example, if the mouse cursor is moved on the voice reproduction button 16 and clicked, the reading of the text part is started. As described above, when the audio playback button 16 is clicked, the client terminal first activates the playback program embedded in the search information and transmitted from the audio conversion server side, and responds from the audio conversion server 10. It operates to download audio data and reproduce it as audio.

(B-5) Voice Reproduction Operation After Voice Reproduction Button Operation FIG. 7 shows a flow of a processing operation after the audio reproduction button 16 is clicked or the like.

First, in step SP11, the sound reproducing means 12 sets a sound number counter to 0.

Subsequently, in step SP12, the display input means 11 transmits a request for a directory name and a sound file name to the sound conversion server 10. At this time, the document search unit 2 checks whether the requested audio file is in the directory.

Here, if a positive result is obtained (if any), the process proceeds to step SP15, and if a negative result is obtained (if no), the process proceeds to step SP13.

Here, it is assumed that there is no voice data corresponding to the number yet, and the process proceeds to step SP13. At this time, the sound reproducing means 12 checks whether or not the document search means 2 has a sound conversion end file. Here, if a positive result is obtained (if any), the process is terminated. If a negative result is obtained (if not), the process proceeds to step SP14, and after waiting for a certain time (the conversion process ends). The process returns to step SP12 again.

The loop processing step SP12-
The processing of SP13-SP14-SP12 is continued until audio data corresponding to the number (in the initial state, audio data corresponding to number 0) is found.

Eventually, a positive result is obtained in step SP12, and the process proceeds to step S15.
Sends audio data to the display input means 11, and the audio reproducing means 12 reproduces the audio data.

Then, in the next step SP16, the audio reproducing means 12 increments the count value of the audio number counter by one, and returns to step SP12 to prepare for the reproduction of the next audio data.

(B-6) Specific Example of Sound Reproducing Operation Finally, a series of operations described above will be described with a specific example.

As shown in FIG. 4, first, the network address of the information that the user wants to search and display is displayed in the window 14.
Enter within. In the case of FIG. 4, "http://www.foo.co.jp/i
ndex.html ".

When the user on the client terminal side who has completed the input presses the line feed key and transmits this address to the voice conversion server 10 side, the document search means 2 searches for the information, and a voice playback button is added to the searched information. 16 and a reproduction program, and transmit to the client terminal.

As a result, the retrieved information 17 and the sound reproduction button 16 are displayed on the display input means 11 of the client terminal as shown in FIG.

On the other hand, the voice conversion server 10 creates a directory where the voice data corresponding to the text data included in the information can be found. For example, create the following directory:

/Audio/http/www.foo.co.jp/index.html/ Under this directory, voiced data is stored for each sentence. For example, this data is given a file name having the following number.

/Audio/http/www.foo.co.jp/index.html/0000.au /audio/http/www.foo.co.jp/index.html/0001.au/audio/http/www. foo.co.jp/index.html/0002.au /audio/http/www.foo.co.jp/index.html/0003.au /audio/http/www.foo.co.jp/index.html/ 0004.au In the example of FIG. 6, voice data of “Welcome to the XX company homepage” is stored in “0000.au”, and “0001.au” is stored.
'' Stores voice data of `` new arrival information '', and `` 0002.
au ”stores voice data of“ Company Profile ”and“ 000 ”
3.au ”stores the voice data of“ employment information ”and“ 0 ”
004.au ”stores voice data of the“ related company ”.

When all text data has been converted to voice data, a file indicating the end of the conversion is created, for example, as shown below.

[0061] /audio/http/www.foo.co.jp/index.html/e
nd.au In this state, as shown in FIG. 6, when the audio reproduction button 16 displayed on the screen is operated, the audio reproduction means 12 firstly outputs “/ audio / http / www. foo.c
o.jp/index.html/0000.au "is asked to the document search means 2.

If the file has not been created yet, it is assumed that "/audio/http/www.foo.co.jp/in
dex.html / end.au ”is asked to the document search means 2.

If this file has not been created, a standby process is performed for a while, and then “/ audio / http / www.
It asks the document search means 2 whether there is a file "foo.co.jp/index.html/0000.au". If there is, the file is received and the sound is reproduced. When playback ends, the next file "/audio/http/www.foo.co.j
Ask if there is a file called "p / index.html / 0001.au" and play it.

By repeating these processes, “/ audio / http /
When the playback of `` www.foo.co.jp/index.html/0004.au '' is completed, `` /audio/http/www.foo.co.jp/index.html/0005.a
The file "u" does not exist, and "/audio/http/www.fo
o.co.jp/index.html/end.au ”, the process ends.

(B-6) Effect of First Embodiment As described above, according to the first embodiment, a document search having a configuration in which a plurality of client terminals are connected to a server terminal via a network. In the system, a text-to-speech conversion function common to all client terminals is provided on the server terminal side, and each client terminal
Since the audio data after the audio conversion is read from the server terminal and played back as needed, the load on the client terminal during the audio conversion can be reduced.

Further, this eliminates the need to install a text-to-speech conversion function in each terminal every time a new client terminal is connected to the network.

Further, when the performance of the text-to-speech conversion function is improved, it is not necessary to perform re-installation for all client terminals, which is advantageous in terms of maintenance management.

According to the first embodiment, the audio data after conversion is stored in the server terminal. Therefore, when the audio data previously issued by another client terminal is stored, Since the desired audio data can be reused without executing the audio conversion process again from the beginning, the utilization efficiency of the entire system can be improved and the time required for reproduction can be shortened.

Further, according to the first embodiment, the text data included in the document is divided into sentences and converted to text-to-speech, and each of them is given a file name and stored. Thus, without waiting for the completion of the voice conversion for all the text data in the document, the voice can be reproduced while executing the voice conversion processing up to the point where the conversion has been completed. The required time can be reduced.

Further, according to the first embodiment, the audio reproduction software common to all the client terminals is stored in the server terminal, and if necessary (for example, when a document search is instructed), the client terminal With the configuration for downloading to the client terminal, it is not necessary to install a reproduction program in advance in each of the client terminals.

(C) Second Embodiment (C-1) Configuration of System According to Second Embodiment FIG. 8 is a block diagram showing the functional configuration of a document search system according to the second embodiment.

The difference between the structure of the document search system according to the second embodiment and the structure of the document search system described in the first embodiment is as shown in FIG.
As can be seen from FIG. 8 with the same reference numerals, the voice conversion server 20 (the voice conversion server 10 without the voice load program storage means 7) is used, and the voice reproduction software is stored in advance in each client terminal. And a voice loading means 21 for reproducing voice data from the voice conversion server 20 based on the software.

In the case of the document search means 2 'in the voice conversion server 20, unlike the case of the voice conversion server 10 described in the first embodiment, a voice playback button and a playback program are added to the searched document. Another difference is that no additional function is provided. This is because in the case of the present embodiment, these functions are provided on each client terminal side.

In the case of the display input means 11 'in the second embodiment, unlike the case of the display input means 11 described in the first embodiment, the function of reading the audio data from the audio conversion server 20 is different. Not provided. This is because the function is separately provided in the voice loading means 21.

The above is the configuration of the document search system according to the second embodiment.

(C-2) Example of Initial Display Screen Next, FIG. 9 shows a basic layout configuration of a display screen displayed on the screen of the display input means 11 'of the client terminal according to the second embodiment. FIG. 9 shows the same parts as in FIG. 4 with the same reference numerals, and their layout configuration is basically the same.

However, in the case of the second embodiment,
Since the software necessary for downloading the audio data from the audio conversion server 20 is stored on the client terminal side, the audio reproduction button 18 is displayed in the window 14 even before the search.

(C-3) Voice Conversion Operation FIG. 10 illustrates a voice conversion operation by the document search system according to the second embodiment. FIG. 10 shows the same or corresponding parts as those in FIG. 5 with the same or corresponding reference numerals.

As can be seen by comparing FIG. 10 with FIG. 5, the operation is the same as that of the first embodiment except that the processing in step SP3A is different.

That is, first, it is assumed that a user using a certain client terminal uses the display input means 11 to input an address of information desired to be searched and displayed.
The address input in step SP1 is sent to the document search means 2 through the terminal control means 5.

In the following step SP2, the document search means 2 'operates to search for information corresponding to the address. However, if the corresponding address does not exist in the same server, the information is transmitted via the network connection means 1. To search for information of the address to be searched using the other document search means 8 and 9, and receive the result.

At step SP3A, the document search means 2 '
Outputs the searched document to the client terminal that issued the search request.

In the next step SP4, the document search means 2
Confirms whether there is voice data corresponding to the searched address in the voice storage means 3. Here, if a positive result is obtained and it is confirmed that the user already has the information, the processing ends.

On the other hand, if a negative result is obtained, the process proceeds to step SP5, where the document search means 2 sends the searched information to the sentence division means 6. Here, the sentence division unit 6 extracts only the text part of the information, performs a process of dividing the extracted text data into one sentence at a time, and returns it to the document search unit 2.

When the division processing is completed, step S
Proceeding to P6, the document search means 2 sends the divided text data to the voice conversion means 4. The voice conversion means 4 converts the received text data into voice data and returns it to the document search means 2.

At step SP7, the document search means 2 passes the voice data to the voice storage means 3. The voice storage means 3
Store the received audio data in the directory. The location of the directory is uniquely determined by the searched address. The audio data is filed for each sentence, and is assigned a sequential number.

At step SP8, the voice storage means 3
Creates an audio conversion end file in the directory indicating that all the audio files have been stored, and ends a series of audio conversion operations.

(C-4) Example of Display Screen after Voice Conversion FIG. 11 shows an example of a display screen appearing on the display screen of the client terminal that has requested information retrieval after the above-described voice conversion. FIG. 11 shows the same parts as in FIG. 9 with the same reference numerals.

The difference between FIG. 11 and FIG. 9 is that FIG.
Information (including text) corresponding to the entered address
17 is only displayed in the window 15.

Here, for example, if the mouse cursor is moved on the sound reproduction button 18 and clicked, the reading of the text displayed in the window 15 is started.

When the audio reproduction button 18 is clicked, the client terminal activates the reproduction program stored in the audio loading means 21, downloads the corresponding audio data from the audio conversion server 20, and reproduces it as audio. Works as follows.

(C-5) Voice Reproduction Operation After Voice Reproduction Button Operation This operation is the same as in the first embodiment.
It is executed according to the procedure shown in FIG. Therefore, the description is omitted.

(C-6) Specific Example of Sound Reproducing Operation Finally, a series of operations described above will be described with a specific example.

As shown in FIG. 9, first, the network address of the information that the user wants to search and display is displayed in the window 14.
Enter within. In the case of FIG. 9, “http://www.foo.co.jp/i
ndex.html ".

When the user on the client terminal side who has completed the input presses the line feed key and transmits this address to the voice conversion server 20 side, the document retrieval means 2 'retrieves the information and transmits it to the client terminal.

As a result, the retrieved information 17 is displayed on the display input means 11 'of the client terminal as shown in FIG.

On the other hand, the voice conversion server 20 creates a directory where the voice data corresponding to the text data included in the information can be found. For example, create the following directory:

/Audio/http/www.foo.co.jp/index.html/ Under this directory, data vocalized for each sentence is stored. For example, this data is given a file name having the following number.

/Audio/http/www.foo.co.jp/index.html/0000.au/audio/http/www.foo.co.jp/index.html/0001.au/audio/http/www. foo.co.jp/index.html/0002.au /audio/http/www.foo.co.jp/index.html/0003.au /audio/http/www.foo.co.jp/index.html/ 0004.au In the example of FIG. 11, voice data of “Welcome to the XX company homepage” is stored in “0000.au”, and “0001.a” is stored.
“u” stores voice data of “What's New” and “000”
2.au ”stores the voice data of“ Company Profile ”and“ 0
003.au ”stores audio data of“ employment information ”
“0004.au” stores voice data of “related company”.

When all the text data has been converted into audio data, a file indicating the end of the conversion is created, for example, as shown below.

/Audio/http/www.foo.co.jp/index.html/end.au In this state, the audio playback button 18 displayed on the screen is operated as shown in FIG. First, the audio loading means 21 first reads “/audio/http/www.foo.co.j
Ask the document search means 2 'if there is a file "p / index.html / 0000.au".

Here, if the file has not been created yet, it is assumed that "/audio/http/www.foo.co.jp/in
dex.html / end.au ”is asked to the document search means 2.

If this file has not been created, after waiting for a while, "/ audio / http / www.
The document search means 2 'is asked whether there is a file "foo.co.jp/index.html/0000.au". If there is, the file is received and the sound is reproduced. When playback ends, the next file "/audio/http/www.foo.c
o.jp/index.html/0001.au ”, and play it back.

By repeating these processes, “/ audio / http /
When the playback of `` www.foo.co.jp/index.html/0004.au '' is completed, `` /audio/http/www.foo.co.jp/index.html/0005.a
The file "u" does not exist, and "/audio/http/www.fo
o.co.jp/index.html/end.au ”, the process ends.

(C-6) Effects of the Second Embodiment As described above, according to the second embodiment, the first embodiment is executed except that the audio reproduction software is held on the client terminal side. The configuration is the same as that of the embodiment, and the same effect as that of the first embodiment can be obtained.

(D) Third Embodiment (D-1) System Configuration According to Third Embodiment FIG. 12 is a block diagram showing a functional configuration of a document search system according to the third embodiment.

The differences between the document search system according to the third embodiment and the document search system described in the first embodiment are the same as those in FIG.
As can be seen from FIG. 11 with the same reference numerals, a speech conversion server 30 (provided with a sentence storage means 31 in place of the speech storage means 3 of the speech conversion server 10) is used.

Here, the sentence storage means 31 is the sentence dividing means 6
This is means for storing text data delimited for each sentence for a certain period of time. Therefore, the text-to-speech conversion is performed each time the sound reproduction is requested.

(D-2) Example of Initial Display Screen The basic layout configuration of the display screen displayed on the screen of each display input unit 11 of the client terminal according to the third embodiment is described in the first embodiment. The configuration is the same as that shown in FIG.

(D-3) Text File Storage Operation FIG. 13 illustrates a text file storage operation by the document search system according to the third embodiment. Note that FIG.
Are the same as those in FIG.

As can be seen by comparing FIG. 13 and FIG. 5, the processing in steps SP4A, SP6A and SP7A is different, and the processing in step SP8 is different.
The operation is the same as that of the first embodiment except that it does not have the processing of FIG.

The operation of storing a text file in the third embodiment will be described below with reference to FIG.

First, it is assumed that a user using a certain client terminal has input an address of information to be searched and displayed by using the display input means 11. The address input in step SP1 is sent to the document search means 2 through the terminal control means 5.

In the following step SP2, the document search means 2 operates to search for information corresponding to the address. However, if the corresponding address does not exist in the same server, the information is transmitted via the network connection means 1. It connects to the network, searches for information of the address to be searched using the other document search means 8 and 9, and receives the result.

In step SP3, the document search means 2
The voice reproduction program and buttons stored in the voice loading program storage means 7 are embedded in the document obtained as a result of the search, and the data is output to the client terminal which issued the search request via the terminal control means 5. .

At the next step SP4A, the document search means 2 checks whether there is text data corresponding to the searched address in the sentence storage means 31. Here, if a positive result is obtained and it is confirmed that the user already has the information, the processing ends.

On the other hand, if a negative result is obtained, the process proceeds to step SP5, where the document search means 2 sends the searched information to the sentence division means 6. Here, the sentence division unit 6 extracts only the text part of the information, performs a process of dividing the extracted text data into one sentence at a time, and returns it to the document search unit 2.

When the division processing is completed, step S
Proceeding to P6A, the document search means 2 sends the divided text data to the sentence storage means 31. The sentence storage means 31 stores the received text data in a corresponding directory.

The directory is uniquely determined by the searched address. Text data is filed for each sentence, and the files are numbered and stored in order. After all the files have been stored, the process proceeds to the next step SP7A.

At step SP7A, the statement storage means 31
Creates a text end file indicating that all of the text files have been stored, and stores it in the directory. Thereafter, a series of text storage operations ends.

(D-4) Example of Display Screen after Storing Text File Also in the case of the third embodiment, the display screen that appears on the display screen of the client terminal that has requested the information search is the layout of the first embodiment. 6, that is, the same as FIG. 6.

(D-5) Voice Reproduction Operation After Voice Reproduction Button Operation FIG. 14 shows a flow of a processing operation after the audio reproduction button 18 is clicked or the like. FIG. 14 is the same as FIG. 7, and the same or corresponding parts are denoted by the same or corresponding reference numerals, except that the processing contents of step SP12A and step SP13A are different, and that step SP17 is included. It has a similar configuration.

Hereinafter, the operation will be described with reference to FIG.

First, in step SP11, the audio reproducing means 12 sets the audio number counter to 0.

At the subsequent step SP12A, the display input means 11 transmits a request for a directory name and a text file name to the voice conversion server 30. At this time,
The document search means 2 checks whether the requested text file is in a directory.

Here, if a positive result is obtained (if any), the process proceeds to step SP17, and if a negative result is obtained (if no), the process proceeds to step SP13A.

Here, it is assumed that there is no text data corresponding to the number yet, and the process proceeds to step SP13A. At this time, the sound reproducing means 12 checks whether or not the document search means 2 has a text end file.
Here, if a positive result is obtained (if any), the process is terminated. If a negative result is obtained (if not), the process proceeds to step SP14, and after waiting for a certain time (the conversion process is terminated). The process returns to step SP12 again.

This loop processing step SP12A
The processing of -SP13A-SP14-SP12A is continued until text data corresponding to the number (in the initial state, text data corresponding to number 0) is found.

Eventually, a positive result is obtained in step SP12A, and when the process proceeds to step 17, the document search means 2 gives the text data to the voice conversion means 4 and converts it into voice data.

When voice data is obtained, the next step S
Proceeding to P15, the document search means 2 sends the audio data to the display input means 11, and the audio reproduction means 12 reproduces the audio data.

Then, in step SP16, the audio reproduction means 12 increments the count value of the audio number counter by one, and returns to step SP12 to prepare for reproduction of the next audio data.

(D-6) Specific Example of Sound Reproducing Operation Finally, a series of operations described above will be described with a specific example.

As shown in FIG. 4, first, the network address of the information that the user wants to search and display is displayed in the window 14.
Enter within. In the case of FIG. 4, "http://www.foo.co.jp/i
ndex.html ".

When the user on the client terminal side who has completed the input presses the line feed key and transmits this address to the voice conversion server 30 side, the document search means 2 searches for the information, and a voice playback button is added to the searched information. 16 and a reproduction program, and transmit to the client terminal.

As a result, the retrieved information 17 and the sound reproduction button 16 are displayed on the display input means 11 of the client terminal as shown in FIG.

On the other hand, the voice conversion server 30 creates a directory where the text data included in the information can be found. For example, create the following directory:

/Text/http/www.foo.co.jp/index.html/ Text data delimited for each sentence is stored under this directory. For example, a file name of the following number is given to the text data file.

/Text/http/www.foo.co.jp/index.html/0000.txt /text/http/www.foo.co.jp/index.html/0001.txt / text / http / www. foo.co.jp/index.html/0002.txt /text/http/www.foo.co.jp/index.html/0003.txt /text/http/www.foo.co.jp/index.html/ 0004.txt In the example of FIG. 6, the text data of “Welcome to the XX company homepage.” Is stored in “0000.txt” and “0001.t
xt ”stores the text data of“ What's New ”and“ 00
02.txt "contains the text data of" Company Profile "
“0003.txt” stores the text data of “employment information”, and “0004.txt” stores the text data of “related company”.

When all text data has been stored,
For example, a file indicating the end of storage as described below is created.

/Text/http/www.foo.co.jp/index.html/end.txt In this state, when the sound reproduction button 16 displayed on the screen is operated as shown in FIG. Voice reproduction means 12
First, "/text/http/www.foo.co.jp/index.h
Ask the document search means 2 if there is a file called "tml / 0000.txt".

Here, assuming that the file has not been created yet, "/text/http/www.foo.co.jp/ind
ex.html / end.txt ”is asked to the document search means 2.

If this file has not been created, after waiting for a while, "/text/http/www.f
It asks the document search means 2 if there is a file "oo.co.jp/index.html/0000.txt". If there is, the text file is given to the voice conversion means 4 to perform voice conversion. The voice file obtained by the voice conversion is given to the client terminal which has issued the search request via the terminal control means 5, and is reproduced by the voice reproduction means 12 of the terminal.

When the reproduction is completed, it is asked whether there is a file "/text/http/www.foo.co.jp/index.html/0001.txt" which is the next file.

By repeating these processes, “/ text / http / w
ww.foo.co.jp/index.html/0004.txt '' is played, then `` /text/http/www.foo.co.jp/index.html/0005.tx
There is no file named `` t '' and `` /text/http/www.foo.
co.jp/index.html/end.txt "is found, and the process ends there.

(D-6) Effect of Third Embodiment As described above, according to the third embodiment, a document search having a configuration in which a plurality of client terminals are connected to a server terminal via a network. In the system, a text-to-speech conversion function common to all client terminals is provided on the server terminal side, and each client terminal
Since the audio data after the audio conversion is read from the server terminal and played back as needed, the load on the client terminal during the audio conversion can be reduced.

[0146] This can eliminate the need to install a text-to-speech conversion function in each terminal every time a new client terminal is connected to the network.

Further, when the performance of the text-to-speech conversion function is improved, it is not necessary to perform re-installation for all client terminals, which is advantageous in terms of maintenance management.

Further, according to the third embodiment, the text data of the retrieved document is stored in the server terminal in units of one sentence. Therefore, the text data previously issued by the instruction of another client terminal is stored in the server terminal. When the text data is retained, desired text data can be reused without executing the text data division process again from the beginning, so that the utilization efficiency of the entire system can be improved and the time required for reproduction can be reduced.

Further, according to the third embodiment, the text data included in the document is subjected to text-to-speech conversion in units of one sentence, and the completion of the voice conversion for all the text data in the document is waited for. Even if it is not, the sound can be reproduced while executing the sound conversion processing up to the point where the division is completed, and the time required until the start of the reproduction can be shortened.

Further, according to the third embodiment, the audio reproduction software common to all the client terminals is stored in the server terminal, and if necessary (for example, when a search for a document is instructed), With the configuration for downloading to the client terminal, it is not necessary to install a reproduction program in advance in each of the client terminals.

Further, according to the third embodiment, only the text data is stored and the voice data is not stored, so that the disk capacity required for the voice conversion server 30 can be reduced.

(E) Fourth Embodiment (E-1) Configuration of System According to Fourth Embodiment FIG. 15 is a block diagram showing a functional configuration of a document search system according to a fourth embodiment. FIG. 15 corresponds to FIGS. 8 and 12 and corresponds to the same part, and is denoted by the same reference numeral. As can be seen from FIG. 15, the functional configuration of the document search system according to the fourth embodiment is based on the second embodiment, and its voice storage means 3 is different from that of the third embodiment. This is replaced with the sentence storage means 31.

That is, the configuration is the same as that of the second embodiment except that the voice conversion server 40 having the sentence storage means 31 is provided. Therefore, the document retrieval means 2 'in this embodiment is not provided with a function of attaching a sound reproduction button or a reproduction program to a retrieved document and sending it to the client terminal.

(E-2) Example of Initial Display Screen In the case of the client terminal according to the fourth embodiment, the function of reproducing the audio data from the audio conversion server 40 is performed by the audio loading means 21 in each client terminal. Since they are stored, the basic layout of the display screen displayed on the screen of the display input means 11 'is as shown in FIG.
That is, this is the same as the case of the second embodiment.

(E-3) Text File Storage Operation FIG. 16 illustrates a text file storage operation by the document search system according to the fourth embodiment. Note that FIG.
Are the same as those in FIG. 13, and the corresponding portions are denoted by the same reference numerals.

As can be seen by comparing FIG. 16 with FIG. 13, the operation is the same as that of the third embodiment except that the processing in step SP3A is different.

Hereinafter, the text file storing operation in the fourth embodiment will be described with reference to FIG.

First, it is assumed that a user using a certain client terminal inputs an address of information to be searched and displayed by using the display input means 11 '. The address input in step SP1 is sent to the document search means 2 through the terminal control means 5.

In the following step SP2, the document search means 2 'operates to search for information corresponding to the address. However, if the corresponding address does not exist in the same server, the information is transmitted through the network connection means 1. To search for information of the address to be searched using the other document search means 8 and 9, and receive the result.

At step SP3A, the document search means 2 '
Outputs the searched document to the client terminal that issued the search request.

In the next step SP4A, the document search means 2 'checks whether there is text data corresponding to the searched address in the sentence storage means 31. Here, if a positive result is obtained and it is confirmed that the user already has the information, the processing ends.

On the other hand, if a negative result is obtained, the flow advances to step SP5, where the document search means 2 'sends the searched information to the sentence division means 6. Here, the sentence dividing means 6 extracts only the text part of the information, performs processing of dividing the extracted text data into one sentence at a time, and returns it to the document search means 2 '.

When the division processing is completed, step S
Proceeding to P6A, the document search means 2 'sends the divided text data to the sentence storage means 31. The sentence storage means 31 stores the received text data in a corresponding directory.

The directory is uniquely determined by the searched address. Text data is filed for each sentence, and the files are numbered and stored in order. After all the files have been stored, the process proceeds to the next step SP7A.

In the step SP7A, the sentence storage means 31
Creates a text end file indicating that all of the text files have been stored, and stores it in the directory. Thereafter, a series of text storage operations ends.

(E-4) Example of display screen after storing text file After this text file storing process, the document for which the search instruction has been issued is displayed on the display screen of the client terminal. This screen is the same as FIG.

(E-5) Voice Reproduction Operation After Voice Reproduction Button Operation This operation is the same as in the third embodiment.
It is executed according to the procedure shown in FIG. Therefore, the description is omitted.

(E-6) Specific Example of Sound Reproducing Operation Finally, a series of operations described above will be described with a specific example.

As shown in FIG. 9, first, the network address of the information that the user wants to search and display is displayed in the window 14.
Enter within. In the case of FIG. 9, “http://www.foo.co.jp/i
ndex.html ".

When the user on the client terminal side who has completed the input presses the line feed key and transmits this address to the voice conversion server 40 side, the document search means 2 'searches for the information, and the searched information is searched for. Is sent to the requesting client terminal.

As a result, the retrieved information 17 is displayed on the display input means 11 'of the client terminal as shown in FIG.

On the other hand, the voice conversion server 40 creates a directory where the text data included in the information can be found. For example, create the following directory:

Under this directory, text data delimited by one sentence is stored under /directory/text/http/www.foo.co.jp/index.html/. For example, a file name of the following number is given to the text data file.

/Text/http/www.foo.co.jp/index.html/0000.txt /text/http/www.foo.co.jp/index.html/0001.txt / text / http / www. foo.co.jp/index.html/0002.txt /text/http/www.foo.co.jp/index.html/0003.txt /text/http/www.foo.co.jp/index.html/ 0004.txt In the example of FIG. 11, “0000.txt” stores the text data of “Welcome to the XX company homepage.”
1.txt "stores the text data of"What's New "
"0002.txt" stores the text data of "Company Profile", "0003.txt" stores the text data of "Employment Information", and "0004.txt" stores the text data of "Related Company" You.

When all the text data has been stored,
For example, a file indicating the end of storage as described below is created.

/Text/http/www.foo.co.jp/index.html/end.txt In this state, when the audio reproduction button 18 displayed on the screen is operated as shown in FIG. Voice loading means 2
First, "/text/http/www.foo.co.jp/inde
x.html / 0000.txt "is asked to the document search means 2 '.

Assuming that the file has not been created yet, "/text/http/www.foo.co.jp/ind
ex.html / end.txt "to the document search means 2 '.

If this file has not been created, a standby process is performed for a while, and then “/text/http/www.f
oo.co.jp/index.html/0000.txt "is asked to the document search means 2 '. If there is, the text file is given to the voice conversion means 4 to perform voice conversion. The voice file obtained by the voice conversion is given to the client terminal which has issued the search request via the terminal control means 5, and is reproduced by the voice reproduction means 12 of the terminal.

When the reproduction is completed, it is asked whether there is a file “/text/http/www.foo.co.jp/index.html/0001.txt” which is the next file.

By repeating these processes, “/ text / http / w
ww.foo.co.jp/index.html/0004.txt '' is played, then `` /text/http/www.foo.co.jp/index.html/0005.tx
There is no file named `` t '' and `` /text/http/www.foo.
co.jp/index.html/end.txt "is found, and the process ends there.

(E-6) Effects of the Fourth Embodiment As described above, according to the fourth embodiment, the third embodiment is implemented except that the audio reproduction software is held on the client terminal side. The configuration is the same as that of the embodiment, and the same effect as that of the third embodiment can be obtained.

(F) Other Embodiments (F-1) In each of the above-described embodiments, a file (voice data file or text data file) is stored in the voice storage unit 3 or the sentence storage unit 31. When
Although a sequential number indicating the playback order is given to the file name, symbols other than numerals may be used for the file name as long as the order in which the files should be played back can be identified.

(F-2) In each of the above embodiments, the sentence dividing means 6 is used to divide the text data included in the retrieved document into units of one sentence. It may be divided or it may be divided in paragraph units. Similarly, in each embodiment, audio data is reproduced in units of one sentence, but may be divided in units of a plurality of sentences or in units of paragraphs.

(F-3) In each of the above embodiments, the case where external document search means 8 and 9 are connected to the document search means 2 or 2 'in the server terminal via a network. As described above, all of these may exist on the same server terminal.

(F-4) In each of the above embodiments, the case where the number of client terminals connected to the server terminal is three and the number of other document search means is two is described. However, neither is limited to this, and any number may be connected.

(F-5) In the first and second embodiments described above, if the addresses on the network are different,
Although the corresponding audio data file is stored in a different directory, the directory to be stored is not limited to this as long as it can be identified which address the audio data file corresponds to.

Similarly, in the above-described third and fourth embodiments, if the addresses on the network are different, the corresponding text data files are stored in different directories. The directory to be stored is not limited to this as long as it can be identified as corresponding to.

(F-6) In the first and third embodiments, as shown in FIG.
Although the sound reproduction button 16 is displayed in the part before the number 7, the display position is not limited to this, and may be displayed after the end of the information 17, for example.

Similarly, in the second and fourth embodiments, as shown in FIG. 9, the window 14 used for inputting an address and the audio playback button 18 are displayed outside the window 15 window. The display position is not limited to this, and may be anywhere on the screen.

[0190]

As described above, according to the present invention, in a document search system in which a plurality of client terminals and a server terminal are connected via a network, text data included in the extracted information is converted to voice. With the configuration in which the voice conversion means for converting data is provided in the server terminal, it is not necessary to provide the voice conversion means in each of the client terminals as in the conventional system.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a document search system according to a first embodiment.

FIG. 2 is a table showing a relationship between embodiments.

FIG. 3 is a block diagram illustrating a configuration example of a voice conversion unit.

FIG. 4 is a diagram illustrating an example of an initial screen of a display input screen (part 1).

FIG. 5 is a flowchart showing a series of operations from a search instruction to the end of a voice conversion operation (part 1).

FIG. 6 is a diagram illustrating a state of a display input screen after predetermined information is read (part 1).

FIG. 7 is a flowchart showing an audio data reproducing procedure (part 1).

FIG. 8 is a block diagram illustrating a configuration of a document search system according to a second embodiment.

FIG. 9 is a diagram illustrating an example of an initial screen of a display input screen (part 2).

FIG. 10 is a flowchart showing a series of operations from a search instruction to the end of a voice conversion operation (part 2).

FIG. 11 is a diagram illustrating a state of a display input screen after predetermined information is read (part 2).

FIG. 12 is a block diagram illustrating a configuration of a document search system according to a third embodiment.

FIG. 13 is a flowchart showing a series of operations from a search instruction to completion of storage of text data (part 1).

FIG. 14 is a flowchart showing an audio data reproducing procedure (part 2).

FIG. 15 is a block diagram illustrating a configuration of a document search system according to a fourth embodiment.

FIG. 16 is a flowchart showing a series of operations from a search instruction to completion of storage of text data (part 2).

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Network connection means, 2, 2 ', 8, 9 ... Document search means, 3 ... Voice storage means, 4 ... Voice conversion means, 5 ... Terminal control means, 6 ... Sentence division means, 7 ... Voice loading program storage means ..., 10, 20, 30, 40 ... sound conversion server, 11, 11 '... display input means, 12 ... sound reproduction means, 13, 14, 15 ... windows, 16, 18 ... sound reproduction buttons, 17 ... information, 21 ... Voice loading means, 31
... Sentence storage means.

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Hideki Yamamoto 1-7-12 Toranomon, Minato-ku, Tokyo Oki Electric Industry Co., Ltd. (72) Inventor Toshiki Murata 1-7-112 Toranomon, Minato-ku, Tokyo Inside Electric Industry Co., Ltd. (72) Inventor Hiromichi Hayashi 3-192-2 Nishi-Shinjuku, Shinjuku-ku, Tokyo Japan Inside Telegraph and Telephone Co., Ltd. Inside Nippon Telegraph and Telephone Corporation (72) Inventor Ikuo Namiki 3-19-2 Nishi Shinjuku, Shinjuku-ku, Tokyo Japan Within Telegraph and Telephone Corporation (72) Inventor Masami Ujiie 20-1, Sakuraoka-cho, Shibuya-ku, Tokyo NTT Inside Electronics Co., Ltd. (72) Inventor Naoko Kato 20-1 Sakuragaoka-cho, Shibuya-ku, Tokyo Inside NTT Electronics Corporation

Claims (10)

[Claims] 1. A document search system in which a plurality of client terminals and a server terminal are connected via a network, wherein the server terminal retrieves necessary information from an address on the network instructed by the search. Voice conversion means for converting text data included in the retrieved information into voice data; voice storage means for storing the voice data; and transmitting the retrieved information to the client terminal which has issued the search instruction. A document search system comprising a terminal control unit. 2. The document search system according to claim 1, wherein the server terminal divides text data included in the extracted information by a predetermined unit, and transmits the divided text data to the voice conversion unit. A document search system comprising text division means for giving a document. 3. The document search system according to claim 2, wherein the server terminal is configured to output the voice when text playback included in the retrieved information is instructed from the client terminal that has issued the search instruction. A document retrieval system, wherein voice data divided for each predetermined unit is read from a storage unit, and is sequentially transmitted to the client terminal. 4. The document search system according to claim 1, wherein said server terminal has a voice load program storage means for storing a voice data reproduction program, and said client terminal having issued a search instruction to said client terminal. A document retrieval system which, when sending out the extracted information, embeds a reproduction instruction button and a reproduction program for audio data together with the information and transmits the information. 5. The document retrieval system according to claim 1, wherein the client terminal is configured to load, from a server terminal, audio data corresponding to text data included in the retrieved information. A document search system comprising: 6. A document search system in which a plurality of client terminals and a server terminal are connected via a network, wherein the server terminal retrieves necessary information from an address on the network for which a search instruction has been issued. And text data storage means for storing the text data contained in the retrieved information; and when instructed to reproduce the text data contained in the retrieved information by voice, the text data storage means stores the text data from the text data storage means. A document retrieval system comprising: voice conversion means for reading and sequentially converting the data into voice data; and terminal control means for transmitting the retrieved information to a client terminal that has issued a search instruction. 7. The document search system according to claim 6, wherein the server terminal divides text data included in the extracted information by a predetermined unit, and stores the divided text data in the text data storage means. A document retrieval system comprising a text division unit for providing a document. 8. The document search system according to claim 7, wherein the server terminal is configured to output the voice when text reproduction included in the retrieved information is instructed from the client terminal that has issued the search instruction. A document search system, wherein the conversion means reads voice data voice-converted for each predetermined unit and sequentially transmits the voice data to the client terminal. 9. The document search system according to claim 6, wherein the server terminal has an audio load program storage means for storing an audio data reproduction program, and the client terminal which has issued the search instruction is provided to the client terminal. A document retrieval system which, when sending out the extracted information, embeds a reproduction instruction button and a reproduction program for audio data together with the information and transmits the information. 10. The document search system according to claim 6, wherein the client terminal loads, from a server terminal, voice data corresponding to text data included in the retrieved information. A document search system comprising: