JPH10289092A - Information processing system and information management method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To use a terminal which has excellent portability as a terminal and further to provide various processing functions without damaging operability between a central processing unit and a terminal in an information processing system which is equipped with a terminal (e.g. client) and information processing functions and is provided with a central processing unit (e.g. server) which exchanges information with a terminal. SOLUTION: A central processing unit 1 unitarily manages requests from each terminal A1 to An . When a prescribed item in a voice is sent from a terminal after the terminal side and the unit 1 are connected, the unit 1 outputs information that corresponds to the prescribed item which is inputted from the terminal side in a voice. In each step of the processing, the unit 1 urges the terminal an item to be inputted next in a voice and interactively performs the exchange of voice information with the terminal.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an information processing system and an information management method including a terminal, a central processing unit having an information processing function, and exchanging information with the terminal.

[0002]

2. Description of the Related Art For example, when an information processing system for exchanging information with a remote terminal is constructed, a server-client system is often used as a form of the information processing system. In this method, requests from many terminals are centrally managed by a central processing unit (server) having an information processing function, and corresponding information is provided to the terminals.
According to this method, the terminal side (client side) only needs to have a function of transmitting and receiving data and the like, so that the terminal side can reduce the storage capacity and the number of required programs. Become. Meanwhile, the central processing unit
On the (server) side, programs started in response to requests from a large number of terminals and data to be transmitted / received can be shared and managed and stored, so that there is an advantage that operation and maintenance are easy.

[0003]

By the way, as such an information processing system, a terminal having high mobility (terminal having excellent portability) is used as the terminal, and a user of the terminal can communicate with the server. It is assumed that the terminal is used by moving to a point. However, in the case of constructing such an information processing system, conventionally, the server generally transmits and receives character information. Therefore, a terminal such as a keyboard having keys necessary for transmitting information is provided on the terminal side. A character input device is required, so that it is difficult to reduce the size of the terminal, and there is a limit in improving portability.

In order to solve such a problem, a method of connecting to a server using voice on a terminal side and processing information can be considered. In this case, a device such as a telephone (mobile phone) having a function of transmitting voice can be used for the terminal, and portability can be improved.

[0005] When a terminal such as a telephone is used as a terminal and information is exchanged with a server by voice, it is desired that various processing functions be realized. However, there is a problem that it is difficult at present to realize various processing functions.

That is, for example, there is a problem that a terminal cannot request a server for complicated processing to obtain information from the server. Specifically, for example, when giving a request from a terminal to the server to output a voice message of a predetermined message (message) to a predetermined destination, the voice information input from the terminal includes at least the destination and the message to the destination. Contents and two pieces of information are required. At this time, when the terminal operator utters at once without separating the two pieces of information of the destination and the contents of the message to the destination, it is difficult for the server to separate these two pieces of voice information. In order to achieve the target processing, the operator of the above is aware of the utterance interval so that the two pieces of information of the destination and the contents of the message to the destination can be respectively recognized by the server, There is a restriction that each piece of information must be pronounced accurately, and it is difficult to improve operability.

[0007] As the types of information that can be processed increase, the number of words to be subjected to speech recognition increases. In general, as the number of vocabularies increases, the accuracy of speech recognition decreases, and if erroneous recognition occurs frequently, the same utterance is repeatedly requested from the server, making it difficult for users to handle. is there.

[0008] The present invention provides a terminal (eg, a client)
In an information processing system including a central processing unit (e.g., a server) having an information processing function and exchanging information with a terminal, it is possible to use a highly portable terminal as a terminal. Still another object of the present invention is to provide an information processing system and an information management method capable of realizing various processing functions between a central processing unit and a terminal without impairing operability.

[0009]

In order to achieve the above object, the invention according to claim 1 comprises at least one terminal,
A central processing unit having an information processing function and exchanging information with at least one terminal, wherein a request for the central processing unit is input to the terminal as voice A request input unit that transmits the input voice to the central processing unit; and an information output unit that receives and outputs information from the central processing unit by voice, and the central processing unit has a predetermined Information processing means for performing information processing, storage means for storing data required for information processing, voice recognition means for receiving voice transmitted from the request input means of the terminal, and voice recognition of the received voice, The information managed by the voice recognition unit is managed, and when the voice information recognized by the voice recognition unit is a request to the information processing unit, the information processing unit sends the request to the information processing unit. Request processing means for performing the same processing, and voice synthesis means for converting the information to be transmitted from the central processing unit to the terminal into a voice by voice synthesis and transmitting the generated voice to the terminal. It is characterized by having.

According to a second aspect of the present invention, in the information processing system according to the first aspect, when the terminal is connected to the central processing unit, the request management means of the central processing unit is connected to the terminal. In response to this, when a voice input for specifying the user is prompted in response to the voice input for specifying the user from the request input means of the terminal, the voice recognition means of the central processing unit Operates the user of the terminal with respect to the central processing unit by comparing the feature of the voice input from the request input means of the terminal with the voice feature of the user of the terminal registered in advance. It is characterized by determining whether or not the user has the right.

According to a third aspect of the present invention, in the information processing system according to the second aspect, the request management means comprises:
The terminal prompts the terminal for voice input by the same user a plurality of times, and when the voice input by the same user is input a plurality of times from the request input unit of the terminal, the voice recognition unit is input a plurality of times. And determining whether the user of the terminal has an operation right to the central processing unit based on the feature amount of the voice of the same user.

According to a fourth aspect of the present invention, in the information processing system according to the first aspect, the request management means comprises:
If the target of the request cannot be specified based on the voice input from the request input means of the terminal, the voice synthesizing means makes a voice inquiry again to the request input means of the terminal.

According to a fifth aspect of the present invention, in the information processing system according to the first aspect, the terminal has a function as a telephone.

According to a sixth aspect of the present invention, in the information processing system according to the first aspect, the request input means of the terminal has a storage unit for storing voice data uttered by the user, and the request input means is provided as required. And has a function of storing the voice data uttered by the user in the storage unit, and, if necessary, reading out the voice data stored in the storage unit and transmitting it to the central processing unit. It is characterized by having.

According to a seventh aspect of the present invention, in the information processing system according to the first aspect, the request management means of the central processing unit urges the terminal to input a voice of mail. In response, when there is a voice input about the mail from the request input means of the terminal, the information processing means of the central processing unit transmits the voice data input from the request input means of the terminal to another user of the storage means. It has a function of storing data in a data storage area.

According to an eighth aspect of the present invention, in the information processing system according to the seventh aspect, the information processing means comprises:
When voice data from the predetermined user is stored in a data storage area of another user of the storage means, the content of the voice data is transmitted from the voice synthesis means to a terminal of the other user. It is characterized by having a function of notifying by.

According to a ninth aspect of the present invention, in the information processing system according to the first aspect, the request management means of the central processing unit prompts the terminal to input a voice about a user's schedule. In response to this, when there is a voice input about the schedule from the request input means of the terminal, the information processing means of the central processing unit writes the contents of the schedule voice input from the request input means of the terminal into a date. And a function of storing audio data in the storage means in correspondence with the time.

The invention according to claim 10 is the invention according to claim 9.
5. The information processing system according to claim 1, wherein the information processing unit has a function of transmitting the content of the schedule, date, and time stored in the storage unit as audio data to the terminal by voice using the voice synthesis unit. It is characterized by doing.

The invention according to claim 11 is the invention according to claim 9
In the information processing system described above, the information processing means converts the contents of a schedule stored in the storage means as audio data to a date and time stored in association with the contents of the schedule, and From the terminal to the terminal by voice.

The invention according to claim 12 is the first invention.
In the information processing system according to the above, the request management unit prompts a user of the terminal to input a predetermined target voice,
In response to this, when the input voice is transmitted from the terminal, a dictionary for voice recognition corresponding to the target of the input voice from the terminal is selected so that the voice input from the terminal is recognized by the voice recognition means. And causing the voice recognition means to recognize the input voice from the terminal using the selected voice recognition dictionary.

According to a thirteenth aspect of the present invention, there is provided at least one terminal and at least one information processing function.
An information management method in an information processing system comprising: a central processing unit that exchanges information with one terminal; wherein the central processing unit centrally manages requests from each terminal; After the connection with the device is made, when a predetermined item is transmitted by voice from the terminal side, the central processing unit outputs information corresponding to the predetermined item input from the terminal to the terminal by voice. At each stage of this process, the central processing unit prompts the terminal to input the next item to be input by voice, and exchanges information with the terminal by voice in an interactive manner. Features.

The invention according to claim 14 is the first invention.
3. The information management method according to 3, wherein the central processing unit comprises:
After the connection is established with the terminal, the terminal prompts the terminal to input necessary items for confirming the operation right of the terminal user, and in response, the terminal uses the terminal. When the item for confirming the operation right of the user is sent by voice, the feature amount of the voice is compared with the feature amount of the user of the terminal registered in advance, and the operation right of the user of the terminal is confirmed. It is characterized by doing.

[0023] The invention according to claim 15 provides the invention according to claim 1.
4. The information management method according to 4, wherein the central processing unit comprises:
When performing the process of confirming the operation right of the user of the terminal and granting the operation right to the user of the terminal, the terminal prompts the terminal to input a request to the central processing unit by voice, and in response to this, When a request is input by voice from a terminal, information corresponding to the request is transmitted to the terminal by voice.

[0024] Further, the invention according to claim 16 is based on claim 1.
3. The information management method according to 3, wherein the central processing unit comprises:
Prompts the terminal to input the next item to be spoken,
In response to this, when a predetermined item is input by voice from the terminal and the predetermined item is transmitted by voice, a dictionary for speech recognition corresponding to the predetermined item is selected, and the predetermined item input from the terminal is selected. It is characterized in that speech is recognized using the selected speech recognition dictionary.

[0025]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram showing a configuration example of an information processing system according to the present invention. Referring to FIG. 1, the information processing system includes at least one terminal A _{1 to An} (n
.Gtoreq.1) and a central processing unit 1 having an information processing function and exchanging information with at least one terminal.

Here, at least one terminal, for example, A
_When information (for example, a request) to the central processing unit 1 is input as a sound, the input sound is
And an information receiving unit 12 for receiving information from the central processing unit 1 by voice.

[0027] As the terminal A ₁ to A _n having such a configuration,
For example, a telephone (for example, a mobile phone handset)
In this case, the request input means 11 of the terminal includes:
A telephone transmitter or the like can be used. Further, as the information receiving means 12 of the terminal, a telephone receiver or the like can be used.

FIG. 2 shows a more specific configuration example of the terminal. In the example of FIG. 2, the terminal is configured as a telephone, and has a telephone main body 6 and a handset 7. Here, the handset 7 and the telephone body 6 have functions as a request input unit 11 and an information receiving unit 12.

Further, in the example of FIG.
A storage unit 8, and switches S _1, S _2, record button 41,
A reproduction button 42 is provided. Here, the switches S ₁ and S ₂ are opened and closed by operating the recording button 41 and the reproducing button 42. For example, when the recording button 41 is operated (pressed), the switches S ₁ and S ₂ are turned off. both closed, and when the playback button 42 is operated, the switch S ₁ is open, which is the switch S ₂ is closed so.

[0030] When the switch S ₁ is pressed record button 41 is in the closed, in the storage unit 8, the voice data input from the transmitter of the handset 7 is stored,
When the reproduction button 42 is pressed and the switch S ₁ is opened and the switch S ₂ is closed, as described later, when there is an inquiry from the central processing unit 1, the telephone main body 6 stores in the storage unit 8. Can be read and transmitted to the central processing unit 1.

That is, in the configuration example of FIG. 2, the request input means 11 also has a function of storing voice data uttered by the user as necessary and outputting the stored voice data. ing.

[0032] The central processing unit 1 includes a storage unit 21 for storing concentrated data, and information processing means 22 for performing predetermined information processing, request input of one of the terminals of the terminal A ₁ to A _n Request receiving means 23 for receiving the voice transmitted from the means 11, voice recognizing means 24 for recognizing the voice received by the request receiving means 23, and managing the information recognized by the voice recognizing means 24; If the information of the voice recognized by the request 24 is a request to the information processing means 22, the request management means 25 causes the information processing means 22 to perform a process corresponding to the request, and is transmitted from the central processing unit 1 to the terminal. Voice synthesis means 2 for converting information to be converted into voice
6 and an information transmitting means 27 for transmitting the voice synthesized by the voice synthesizing means 26 to the terminal.

Each of the terminals A _{1 to An} and the central processing unit 1 are connected so that information (voice) is transmitted and received through a predetermined line 2, for example. The line 2 may be a wired line such as a public telephone, or may be a wireless (radio wave) line when a mobile phone is used as a terminal. Also, this line 2 is used for information (voice) from a plurality of terminals.
It can be configured as a multiplex line so as to cope with the above.

FIG. 3 is a diagram showing a more specific configuration example of the central processing unit 1. In the example of FIG. 3, in the central processing unit (server) 1, the respective processing means 24, 25, 2
6, 27, 22, and 21 can mutually exchange data using a network 4 such as an Ethernet network. In the example of FIG. 3, the request receiving unit 23 and the information transmitting unit 24 of the central processing unit 1 of FIG. 1 are provided in the line concentrator 3. Here, the line concentrator 3 is used to combine these multiplexed lines into one when the line 2 is configured as a multiplexed line. By putting together a multi-line in concentrator 3, also the terminal of the terminal A ₁ to A _n throat can control the same central processing unit (server) 1.

In the central processing unit 1, the storage means 2
1 is configured to be able to store various types of information and data used in the processing of the central processing unit 1 including information transmitted from a terminal, and a storage device such as a hard disk is used as the storage unit 21. be able to. The storage unit 21 is configured to be able to store various types of information and data for each user of the terminal (in a data storage area for each user).

The information processing means 22 stores the request in the storage means 21 under the control of the request management means 25 when a predetermined request is received from the terminal, and stores the request in the storage means 21. It has various information processing (program) functions to read information and data in response to a request from the terminal, perform processing related to the information requested by the terminal, and output the processing result. ing.

FIG. 4 shows an example of the program structure of the information processing means 22. When the information processing number is given from the request management means 25, the information processing means 22 executes the program corresponding to the information processing number. (Processing).

More specifically, if the request from the terminal requests a voice mail service (a service for transmitting the voice (message) of the user of the terminal to the user of another terminal), the request management means 25 Information processing means 2 under the management of
Reference numeral 2 has a function of storing data (message) input and transmitted by voice from the request input unit 11 of the terminal in a data storage area of another user of the storage unit 21. When data (message) from another user is stored in the data storage area of the user, the information processing means 22 causes the voice synthesizing means 26 to synthesize the data (message) by voice and transmits the information. It has a function of notifying the user of the contents of the data (message) by voice using the means 27.

When the request from the terminal is a schedule management service, under the control of the request management means 25,
The information processing means 22 has a function of adding the date and time of the user's voice (schedule contents) input and transmitted from the request input means 11 of the terminal to the user's data storage area of the storage means 21. have. That is, it has a function (schedule input function) of storing the contents of the user's schedule (future action contents) in association with the date and time. Further, the information processing means 22 has a function of transmitting and outputting the contents of the schedule stored in the storage means 21 and the date and time to the terminal in response to a request from the terminal (schedule confirmation function). have. That is, in response to a request from the terminal, the voice synthesizing unit 26 synthesizes the voice of the data related to the stored schedule, and has a function of notifying the user of the data regarding the schedule by voice using the information transmitting unit 27. ing. Further, the information processing means 22 has a function of transmitting the contents of the schedule stored in the storage means 22 to the terminal on the date and time stored in association therewith (notifying the user). Function).

The voice recognition means 24 is configured to be able to set various recognition dictionaries required for voice recognition. The voice input from the terminal received by the request receiving means 23 is collated with the recognition dictionary for recognition. It is supposed to.

FIG. 5 is a diagram showing a configuration example of the voice recognition means 24. In the example of FIG. 5, the voice recognition unit 24 performs a voice analysis on the voice data received by the request receiving unit 23 and converts the voice data into a feature pattern, and a reference pattern used as a recognition dictionary is stored in a database. And a recognition dictionary setting for selecting a reference pattern stored in the recognition dictionary database 32 and setting it as a recognition dictionary when dictionary selection data is given from the request management means 25. Part 3
3 and a matching unit 34 for matching a feature pattern from the acoustic analysis unit 31 with a recognition dictionary set by the recognition dictionary setting unit 33, that is, a reference pattern, to perform voice recognition.

Here, the analysis in the acoustic analysis unit 31 includes, for example, a document “Niimi,“ Speech Recognition ”, Kyoritsu Shuppan (197
9) ”and the like. For example, a 10-dimensional cepstrum parameter and a first-order difference between power and power can be extracted and used as a feature pattern. However, the analysis conditions are as follows: sampling frequency: 16 kHz, high-frequency emphasis: primary difference, 256-point Hamming window, update cycle: 10 ms, and LPC analysis order: 20. The analysis method is not limited to the above, and any other acoustic analysis method such as frequency analysis may be used.

The recognition dictionary database 32 has a structure as shown in FIG. 6, for example. That is, in the recognition dictionary database 32, a voice dictionary is registered in advance in correspondence with the voice recognition dictionary number.
For example, the voice recognition dictionary number 1 includes a voice feature pattern "Kato" uttered by a specific user.
(Standard pattern) is registered as a reference pattern. In FIG. 6, the reference pattern column is described in text for convenience, but actually, the reference pattern column is
A pattern obtained by analyzing a voice corresponding to the text is stored.

When the recognition dictionary database 32 has the structure shown in FIG. 6, a speech recognition dictionary number can be used as the dictionary selection data. That is, when a speech recognition dictionary number is given from the request management unit 25, the recognition dictionary setting unit 33 selects a reference pattern corresponding to the speech recognition dictionary number from the recognition dictionary database 32 and sets the recognition dictionary. Can be.

In addition to the function of simply recognizing and outputting an input voice, the voice recognizing means 24 further has a feature amount of the voice input from the request input means 11 of the terminal in accordance with a command from the request managing means 25. Has a function of checking whether or not the user's voice feature amount matches a previously registered user's voice feature amount, and determining whether or not the user has an operation right to the central processing unit 1 (the information processing means 22). I have.

Note that in order to make the determination of the operation right strict, voices by the same user may be input from the request input means 11 of the terminal a plurality of times, and the above-described processing may be repeated. In this case, it is possible to more reliably check whether or not the user has the operation right to the central processing unit 1 (the information processing means 22).

FIG. 7 is a diagram showing a configuration example of the voice synthesizing means 26. In the example of FIG. 7, the speech synthesis unit 26 analyzes the text data for speech synthesis and converts the text data into a symbol expressing the speech in units, and a segment dictionary in which the speech data is stored in syllable units. And a synthesizer 43 for synthesizing the text data analysis result (that is, the speech expression symbol of the text data) from the analyzer 41 based on the unit dictionary 42 to obtain synthesized speech data.

The request management means 25 has program data (starting program) having a structure as shown in FIG. 8, for example. Here, each program data is managed by a number, and in the example of FIG.
For example, by designating the program number 0, the authentication program can be started (the processing of the authentication program can be performed).

The authentication program is a program for judging whether or not the user is a connectable person (a program for confirming whether or not the user has an operation right to the information processing means 22). 25 executes the authentication program, and causes the voice recognition means 24 to confirm the presence or absence of an operation right to the central processing unit 1 (the information processing means 22).

FIG. 9 shows an example of the data structure of the request management means 25 in the case of an authentication program. In the structure shown in FIG. 9, the request management means 25 performs processing on a line indicated by a data number. Here, the speech synthesis text data field stores text contents necessary for generating speech using the speech synthesis means 26, and the speech recognition dictionary number field serves as a recognition target. The voice recognition dictionary number is described.

In the structure shown in FIG.
Starts processing from the row whose data number is 1. The destination data number indicates a data number to proceed to the process currently being processed.

The authentication program execution process (that is, the process of determining whether or not the user has the operation right for the central processing unit 1) is performed, for example, by inputting a voice of a predetermined item (contents) by the same user a plurality of times at the request of the terminal. The voice recognition means 24 recognizes a voice of a predetermined item (content) of the same user, which is input from the means 11 and is input a plurality of times,
This can be performed by collating with a user's voice feature amount registered in advance.

FIG. 10 shows an example of the data structure of the request management means 25 in the case of a menu program. Referring to FIG. 10, the structure of the menu program is the same as the structure of the authentication program of FIG. 9 except that the item of the information processing number is further added. The program of the information processing means 22 can be designated (the information processing means 22 can be started). In the case of this menu program as well, the request management means 25 makes the data number 1
The process is started from the line. The destination data number indicates a data number to proceed to the process currently being processed.

FIG. 11 shows an example of the data structure of the request management means 25 in the case of a voice mail program.

FIG. 12 shows an example of the data structure of the request management means 25 in the case of a schedule management program.

As described above, in the present invention, the central processing unit 1
Is a request from each terminal A ₁ to A _n and centralized management, after the connection between the terminal and the central processing unit 1 has been made, the predetermined items are transmitted by voice from the terminal side, the central processing unit 1 Information corresponding to a predetermined item input from the terminal is output to this terminal by voice, and at each stage of this processing, the central processing unit sends a next item to the terminal to the terminal. Voice prompting is provided, and information is exchanged with the terminal by voice in an interactive manner.

More specifically, after the central processing unit 1 is connected to the terminal (for example, a line connection), the central processing unit 1 requires the terminal to confirm the operation right of the user of the terminal. When the terminal prompts the user to input an item to confirm the operation right of the user of the terminal by voice, the feature amount of the voice and the pre-registered The feature amount of the user is collated and the operation right of the user of the terminal is confirmed.

When the central processing unit 1 performs a process of confirming the operation right of the user of the terminal and recognizes the operation right of the user of the terminal, the central processing unit 1 issues a request (C) to the terminal. (E.g., a mail service or a schedule management service) is prompted by voice, and in response to this, when a request is input by voice from the terminal, information corresponding to the request is transmitted to the terminal by voice. .

The request management means 25 determines which of the above programs is to be designated based on the request (voice request) input from the request input means 11 of the terminal. Terminal input means 11
When it is not possible to specify the target (program) of the request to instruct the information processing means 22 based on the request inputted from the terminal, the inquiry to the request input means 11 of the terminal is made again.

As described above, when there is a request from the terminal, the request management means 25 determines whether the request can be processed, and if the request can be processed, sends the request to the information processing means 22. Then, the information processing unit 22 is caused to perform a process according to the request, and to output a processing result. Since the information as the processing result of the information processing means 22 is obtained as text information, the speech synthesis means 2
Reference numeral 3 designates a configuration in which text information obtained by the information processing unit is voice-synthesized, converted into voice, and transmitted to the terminal via the information transmission unit 27.

On the other hand, when the request management means 25 determines that the request from the terminal cannot be processed,
In order to specify the contents of the request, the terminal is further urged to transmit the request to the terminal via the information transmitting means 27.

The request managing means 25 selects a dictionary to be recognized from various recognition dictionaries provided in the voice recognizing means 24 in accordance with an object to be managed, such as a voice mail or a schedule management, and selects the selected dictionary. It has a function of causing the voice recognition unit 24 to perform recognition processing using only the dictionary that has been set. By providing this function, erroneous recognition in the voice recognition unit 24 can be reduced.

Next, the processing operation of the information processing system having such a configuration will be described.

[0064] Now, the mobile phone is used as a terminal A ₁ ~A _n. In this case, a telephone uses a transmitter as request input means 11 from a remote place, and a receiver as information output means 12. And in this case,
As the line 2, a circuit using radio waves is used, and sound is transmitted using radio waves.

Before the information processing system is actually used, information on the user is registered in advance. For example, the recognition dictionary database 32 of the central processing unit 1 of the speech recognition means 24, for example of the terminals A ₁ to A _n given word to the user (such as name and password) to thereby utterance feature extraction result speech A feature pattern (standard feature pattern) is registered in advance as a reference pattern as shown in FIG.

When the user of the terminal, for example, A ₁ uses the central processing unit 1, first, he uses the function of the terminal A ₁ as a telephone to call the central processing unit 1 (to the central processing unit 1). Dialing), central processing unit (server)
Connect to 1. That is, by dialing the telephone number of the central processing unit (server) 1 from the telephone, the connection with the central processing unit 1 can be established.

When the line connection between the user's terminal A ₁ and the central processing unit 1 is completed, the request management means 25 determines whether the user is a connectable person. First, the authentication program is started. Note that various activation programs are managed by numbers, and in the example of FIG. 8, the request management unit 25 can activate the authentication program by designating the program number 0, for example.

When the authentication program is started, the request management means 25 starts processing from the row of data number 1 in FIG.
That is, first, the voice synthesis unit 26 is requested to generate voice data. Note that the content of the voice data to be generated at this time is described in the column of the voice synthesized text data in FIG. 9, and the content of the voice data corresponding to the data number 1 is
"Please enter your name."

Upon receiving the text data of the content of the voice data to be generated from the request management means 25, the voice synthesizing means 26 converts the text data into a symbol expressing voice in units by the analyzing part 41, and In 43, the analysis unit 4
The voice data corresponding to the syllable converted in step 1 is selected from the segment dictionary 42 and combined.

Thus, for example, when the kanji-kana-maji text data "Please enter a name." Corresponding to data number 1 in FIG. 9 is received from the request management means 25, the speech synthesis means 26 Is converted to a symbol of "Namaeyuryokushitekidasai."
Audio data generated by the voice synthesis section 26 in this manner is transmitted from the information transmitting unit 27 to the terminal (telephone set) A _1, the terminal A _1, from the information output means (handset) 12,
"Please enter your name."
Prompt the user for a name.

[0071] Thus, the user of the terminal A _1, from the request input means (transmitter) 11 of the terminal A _1, enter the name of the self voice. If the name of the user is entered by voice from the request input section 11 of the terminal A _1, the input speech (names)
Is transmitted from the request input unit 11 of the terminal to the central processing unit 1. In the central processing unit 1, when the request receiving unit 23 receives the voice (name) transmitted from the terminal A ₁ ,
The request management unit 25 requests the voice recognition unit 24 to recognize the voice input by the user. Thus, the voice recognition unit 24 recognizes the received voice.

That is, the voice recognizing means 24 operates at the terminal A ₁
The voice (name) transmitted from the
(A feature pattern), and the matching unit 34 matches the voice feature amount (reference pattern) of the user name registered in advance with the information processing unit 22.
Check if this user has the operation right to

More specifically, in this case, the request management means 25 uses the speech recognition dictionary numbers 1, 2, and 3 corresponding to the data number 1 in FIG.
Thus, the recognition dictionary setting unit 33 of the voice recognition unit 24 selects the reference patterns corresponding to the three voice recognition dictionary numbers 1, 2, and 3 from the recognition dictionary database 32 of FIG. Create a dictionary. In this case, a recognition dictionary for recognizing "Kato", "Ito", and "Kojima" is created. Then, when the voice (name) of the user of the terminal A ₁ is input, the voice recognition unit 24 outputs the feature pattern from the acoustic analysis unit 31 and the recognition dictionary set by the recognition dictionary setting unit 33 (that is, “Kato”). ","Ito",
Reference section 3 for recognizing “Kojima” and matching section 3
4, and the score between the input voice feature pattern and the reference pattern is obtained by, for example, the following equation.

[0074]

S _i = T−D (X, W _i )

Here, S _i is a score for the i-th reference pattern. Also, D (X, W _i) is the distance between the feature pattern X and the i-th reference pattern W _i of the input speech, using the Euclidean distance. T is a threshold value and is used to determine a case where the input voice does not belong to any of the candidates. Note that the calculation of the score is not limited to Equation 1, and likelihood may be used like a hidden Markov model.

The collating unit 34 of the voice recognition means 24
Calculates the highest score among the obtained scores S _i (1 ≦ i ≦ I) by the following formula, and sends the reference pattern number to the request management means 25 as a recognition result.

[0077]

(Equation 2)

[0078] Incidentally, in the number 2, I is the number of reference patterns W _i, in the example of the data number 1, I is 3. Further, when all the scores S _i are negative, the voice recognition means 24 does not belong to any candidate and sends a symbol “other” meaning reject to the request management means 25.

Next, the request management means 25 determines the next destination data number according to the recognition result from the voice recognition means 24. For example, when the input voice is recognized as “Kato”, the request management unit 25 receives the voice recognition dictionary number 1 and
, 2 is determined as the destination data number, and the processing of data number 2 in FIG. 9 is started.

The processing for data number 2 is performed in the same manner as the processing for data number 1. In other words, from the central processing unit 1 to the terminal A _1, the voice of "Please enter your password." Is transmitted. Then, the user of the terminal A ₁ is, When you hear a voice saying, "Please enter your password.",
A word registered in advance in the central processing unit 1 as a password for "Kato", for example, "monkey" is input by voice. This input voice is transmitted from the terminal A ₁ of the user to the central processing unit 1, and the voice recognition unit 24 of the central processing unit 1 converts the characteristic pattern of the input voice (password) into “4” of the voice recognition dictionary number 4. It is checked whether or not it is most similar to the reference pattern of “monkey” (whether or not the score with “monkey” is high).

As a result, if the score with “monkey” is high, the request management means 25 determines that the correct password has been input and proceeds to data number 5, otherwise proceeds to data number 6. In this way, by checking the password, it is possible to strictly prevent a person other than the registered user from entering the central processing unit (server) 1. By creating these dictionaries from the registered user's voice, it is possible to design such that even if another user utters "monkey", the collation score is reduced. Therefore, for example, even if another user utters “monkey” after the password is known, the recognition result is “other”.
Thus, the use of the central processing unit (server) 1 is not permitted.

[0082] In this way, the execution processing of the authentication program, provides an indication at each time speech from the central processing unit 1 to the terminal A _1, the user of the terminal A ₁ is the instruction from the central processing unit 1 For example, the process between the terminal A ₁ and the central processing unit 1 can be interactively proceeded such that a response is made by voice, and it is possible to easily confirm whether or not the user has the operation right to the central processing unit 1.

Note that in order to make the determination of the operation right strict, voices from the same user are input a plurality of times from the request input means 11 of the terminal A ₁ , and the above processing is repeated, whereby the central processing unit 1 (information The presence or absence of this user's operation right to the processing means 22) can be more reliably confirmed.

Finally, this processing reaches data number 5 or data number 7 in FIG. When "np" is described in each column of FIG. 9, the process is not performed. If "exit" is described in the destination data number, the program ends. The number (1, 2) following "exit" is a program number, and a program corresponding to this program number is searched for and activated with reference to FIG. Therefore, if it is "exit1," the request management means 25 refers to the program number 1 in FIG. 8, activates the menu program, and ends the authentication program. If it is "exit2", the connection release program is started, and the use of the central processing unit (server) 1 is terminated.

That is, when "exit1" is described in the destination data number, the request management means 25 refers to the program number 1 in FIG.
The program of the information processing means 22 is designated with reference to the information processing number of the menu program structure in FIG. The program of the information processing means 22 is stored in a hard disk or the like in a structure shown in FIG.

More specifically, in the case of the menu program (in the case of FIG. 10) as well as in the case of the authentication program (as in the case of FIG. 9), the request management means 25
The processing of data number 1 is performed. In this data number 1,
In FIG. 10, since "np" is described in the fields of the speech synthesized text data and the speech recognition dictionary number, the speech management processing and the speech recognition processing are not performed, and the request management means 25 is executed.
Requests the information processing means 22 for the processing of the information processing number 1. The information processing means 22 searches for a login as the process of the information processing number 1 from FIG. That is, the login of the user is permitted, and a predetermined work area is secured in the storage unit 21.

Thereafter, the request management means 25 proceeds to the processing of data number 2 in FIG. In the process of data number 2 in FIG. 10, the request management unit 25 causes the voice synthesis unit 26 to synthesize the text data of “Which service do you want to use?” Send. As a result, the terminal A _1, the voice is output from the information output means (receiver) 12, "Are you sure that you want to use which service?", The user of the terminal, in response to this, the service name you want to use Utter.

In the example of the menu program shown in FIG. 10, a voice mail service and a schedule management service are registered as service names. Therefore, the user of the terminal can select one of the voice mail service and the schedule management service. By uttering the service name, any of the voice mail service and the schedule management service can be used.

First, the case where the voice mail service is used will be described. "Which service do you use?"
In response to the voice, the user of the terminal A ₁ is entered by voice "voice mail service", the input speech is transmitted from the terminal A ₁ of the user to the central processing unit 1, a central processing unit 1 Is recognized by the voice recognition means 24, the request management means 25 reaches the data number 3 in FIG.
The voice mail program corresponding to the program number 3 in FIG. 8 is started. Then, the processing of the voice mail program starts from data number 1 in FIG. In FIG. 11, since the information processing number corresponding to the data number 1 is 2, the request management means 25 performs the processing of the information processing number 2 in FIG.
Ask 2

In the column of information processing number in FIG. 11, when R: [] follows the number, it means that the processing result input from the information processing means 22 is substituted for the variable in []. In the column of information processing number in FIG. 11, when S: [] follows the number, the number of the speech recognition result is substituted into a variable in [] and output to the information processing means 22. This means requesting the used information processing.

In the above example, since R: [KENSUU] follows information processing number 2 corresponding to data number 1 in FIG. 11, the information processing means 22 executes the processing for confirming the number of received mails. when mail number (mail number reachable to the user of the terminal a ₁₎ is given to the request management unit 25, which is assigned to [KENSUU] in text format. That is, the mail is stored in the storage unit 21 for each user in, for example, the format shown in FIG. 13. Therefore, the information processing unit 22 stores the user currently stored in the storage unit 21 in the format shown in FIG. 13. by counting the number of mail to a _1, it can be delivered to the total number of request management unit 25 substitutes the received mail number to the user a ₁ to the variable [KENSUU].

Next, since the destination data number corresponding to the data number 1 in FIG.
The processing of data number 2 in FIG. 11 is performed. In the process of the data number 2, the request management unit 25 causes the voice synthesizing unit 26 to perform voice synthesis of "[KENSU] mail has arrived.", And causes the information transmitting unit 27 to transmit this voice to the terminal. Thus, the user of the receiver terminal A _1, can hear the voice informing mail number addressed this user.

Next, the request management means 25 proceeds to the processing of data number 3 in FIG. In the process of data number 3, the request management unit 25 synthesizes the text data of "Do you want to send?" And the text data of "Do you want to receive?" The user of the terminal A ₁ hears these voices from the receiver, utters one of “transmit” and “receive”, and can respond to the central processing unit 1 with voice.

Now, the user of the terminal A ₁ receives, for example, “transmit”
When this is received and recognized by the central processing unit 1, the request management means 25 proceeds to the processing of data number 4 in FIG.

In the processing of data number 4, the request management means 2
5: Text data of "Who are you going to send?"
To the terminal by voice. Accordingly, the terminal A _1, the information output unit (receiver) 12 Voice "Who do you send?" Is outputted from the user terminal, in response thereto, uttering destination (address) . The user utters the address, which is transmitted to the central processing unit 1 and recognized,
The request management means 25 substitutes the address entered by the user into the variable [PERSON] after the information processing number 3 corresponding to the data number 4 in FIG. Then, the information processing means 22 uses this variable PERSON to correspond to the information processing number 3 (the destination specifying process in FIG. 4).
A data area is secured in the work area of (address) in a format as shown in FIG. At the same time, information processing function 2
2 acquires the date from the clock function built in the central processing unit 1 and writes the date data in the date column of the data area of the recipient. Similarly, it acquires user information of the terminal A ₁ that is registered in advance in the storage unit 21, which writes the sender name of the data area.

Next, the request management means 25 proceeds to the data number 8 in FIG. In the process of data number 8 in FIG.
The request management means 25 causes the voice synthesis means 26 to perform voice synthesis on the text data of "Please input the message", and causes the information transmission means 27 to transmit this to the terminal by voice. Thus, the terminal A _1, is the audio output of "Please enter a business' from information output means (receiver) 12, the user of the terminal, in response thereto, a business (message) Utter. The message input from the terminal by voice is transmitted to the central processing unit 1 and received by the central processing unit 1.

At this time, the information processing means 22 of the central processing unit 1 performs the processing of the information processing number 4 in FIG. 4 (the information processing number 4 corresponding to the data number 8 in FIG. 11), that is, the voice recording transmission processing. Then, the received input voice (recorded voice data), that is, the voice data of the message is written in the voice data column of the data area of the above-mentioned recipient.

As described above, the user of the terminal can use the mail transmission service of the central processing unit 1.

On the other hand, in the process of data number 3 in FIG. 11, the request management means 25 synthesizes the text data of "Do you want to send?" And the text data of "Do you want to receive?" In response, the user of the terminal A ₁ utters, for example, “reception” as a response, and when this is received and recognized by the central processing unit 1, the request management means 25 executes the processing of the data number 5 in FIG. Move on to

In the process of data number 5, the request management means 2
Reference numeral 5 indicates that the text data of "Receive mail" is to be synthesized by the voice synthesizing means 26, and this is transmitted to the information transmitting means 27.
To the terminal by voice. As a result, in the terminal A ₁ , the voice of “Receive mail” is output from the information output means (receiver) 12.

In the central processing unit 1, since the information processing number corresponding to the data number 5 in FIG. 11 is 5, the information processing means 22 executes the processing corresponding to the information processing number 5 (the mail reception processing in FIG. 4). ). In the mail receiving process, the data stored in the data area in the format as shown in FIG. 13, that is, the data stored in the date and sender fields are converted into voice data by the voice synthesis means 26. The contents of the mail (voice data) stored in the voice data column are transmitted as voice data to the terminal as it is. Thus, the terminal A _1, from the information output unit (receiver) 12, the date, sender name, audio is output of the contents of the mail, the user of the terminal A _1, together with the contents of the mail, which, when no one You can check if it arrived from. Subsequent processing is performed in the same manner as the mail transmission service processing described above. When the mail receiving service ends in this way, the process returns to the menu program.

[0102] Thus, in the voice mail service, a certain terminal, for example, the user of the A ₁ is other terminals, for example, when sending voice mail to the user of the A ₂ is other input voice data from the terminal A ₁ stored in the data storage area of the user of the terminal a _2. Also, the user of the other terminal A ₂ receives the terminal A _1.
When voice mail directed to the user of is reachable, stores the mail (voice data) from a user of another terminal A ₂ in the data storage area of the terminal A _1, the voice synthesis section 26, information the contents of the mail (voice data) may notify the user of the terminal a ₁ using the transmission unit 27.

Next, a case where the schedule management service is used will be described. In the menu program shown in FIG. 10, "Do you want to use which services?" Of data number 2 in response to the voice of, and the user of the terminal A ₁ is input by voice "schedule management service", the input speech , transmitted from the terminal a ₁ of the user to the central processing unit 1, the request management unit 25 of the CPU 1, 10, reaches the data number 4, the "schedule management service to the terminal a ₁ A message "Starting" is sent to inform that effect, and the schedule management program corresponding to program number 4 in FIG. 8 is started. And FIG.
As in the case of the processing of the voice mail program of FIG.
The process of the schedule management program starts from data No. 1 of No. 2.

In the process of data number 1 in FIG. 12, the request management means 25 asks “Do you want to input a schedule?
Text data of "Do you want to confirm?"
In step 6, the voice is synthesized, and this is transmitted from the information transmitting means 27 to the terminal by voice. As a result, the terminal A _1, "Do you want to enter the schedule?" From the information output means (receiver) 12, voice of "Do you want to check?" Is output, the user of the terminal, in response to this To enter the schedule or confirm the schedule.

The user utters “input of schedule”, which is received by the central processing unit 1,
4, the request management means 25 proceeds to the processing of data number 2 in FIG. In the process of the data number 2, the request management unit 25 causes the voice synthesis unit 26 to perform voice synthesis on the text data of "Please enter a date", and causes the information transmission unit 27 to transmit this to the terminal by voice. Accordingly, the terminal A _1, the information output unit (receiver) 12 voice "Please enter the date" is output from the user terminal, in response thereto, to speak the date.

When the user of the terminal A ₁ utters the date and the date is received by the central processing unit 1, the voice recognition means 24 recognizes the date. The data number 2 in FIG.
In the column of the voice recognition dictionary number corresponding to, the numbers are written in parallel, separated by “,”. This indicates that the reference patterns in FIG. 6 are combined in the arranged order and collated with the input voice. For example, if "16, 28" is written, it indicates that comparison is to be made with a reference pattern "January 1st".

When the date uttered by the user is thus recognized, the information processing number corresponding to data number 2 is 7 in FIG. It requests the information processing means 22 to perform the processing of the information processing number 7, that is, the date recording processing. Thereby, the information processing means 22 performs a date recording process. In the date recording process, the variable [DAT
E], and the voice recognition number of FIG. 6, for example, “16,
Reference is made to the reading column corresponding to "28", and the referred reading (for example, "Ichiga Tsukutachi") is stored in the date column of FIG.

[0108] Next, the request management unit 25 proceeds to a process of the data number 6 in FIG. 12, simultaneously with voice input of the date, to voice input time to the user of the terminal A _1.
When the time is recognized in the central management device 1, the time is stored in the time column of FIG. 14 in the same manner as in the case of the date.

Next, the request management unit 25 proceeds to the processing of the data number 7, and the request management unit 25 causes the voice synthesis unit 26 to perform voice synthesis on the text data of “Please enter a message”, and The information is transmitted from the information transmitting means 27 to the terminal by voice. Thus, the terminal A _1, is the audio output of "Please enter a business' from information output means (receiver) 12, the user of the terminal, in response thereto, a business (message) Utter. The message input from the terminal by voice is transmitted to the central processing unit 1 and received by the central processing unit 1.

At this time, the information processing means 22 of the central processing unit 1 performs the processing of the information processing number 4 in FIG. 4 (the information processing number 4 corresponding to the data number 8 in FIG. 11), that is, the voice recording processing. Then, the received input voice (recorded voice data), that is, the voice data of the business is written in the voice data column of FIG.

Thus, the contents of the schedule are
In the format shown in FIG. 14, the information is stored in the storage unit 21 for each user. That is, the schedule of the user can be stored in the storage unit 21 of the central processing unit 1. Then, at this stage, end the schedule input process (disconnect the line connection)

Thereafter, the information processing means 22 periodically compares the clock built in the central processing unit 1 with the date and time columns in FIG. When the date and time stored in the data area of FIG. 14 have been reached, dial the user's terminal (telephone) using the pre-registered user information, and Connect the telephone line). After that, the data (text data) stored in the date and time fields of the data area in FIG. 14 is converted into voice by the voice synthesizing means 26, and the data stored in the voice data field is left as it is. User's terminal as voice data
(Telephone) to the A _1. Thus, the user of the terminal A ₁ is listening to the content of the schedule to be sent from the central processing unit 1, for example, it is possible to know the schedule of the day.

If the user utters “confirmation” in the processing of data number 1 in FIG. 12, the request management means 25 proceeds to the processing of data number 3 in FIG. In the processing of data number 3, the request management means 25 converts the text data of "
The information is transmitted from the information transmitting means 27 to the terminal by voice. As a result, in the terminal A ₁ , the voice of “confirm the schedule” is output from the information output means (receiver) 12.

The request management means 2 of the central processing unit 1
In 5, the process proceeds to the process of data number 3 in FIG. 12, and since the information processing number corresponding to data number 3 is 10, the process of information process number 10 in FIG. I do. Thereby, the information processing means 22 performs a schedule confirmation process. In the schedule confirmation processing, the data stored in the column of the date and the sender's name in FIG. 14 is converted into a voice by the voice synthesizing means 26 and output to the user's terminal (telephone). Also,
The data stored in the audio data column of FIG. 14 is output as it is to the user's terminal (telephone) as audio data.

Thus, the user of terminal A ₁ can check the schedule by listening to the schedule transmitted from central processing unit 1.

As described above, the schedule of the user, that is, the content of the future action is stored in the form of audio data corresponding to the date and time, and the content of the stored audio data and the date and time are stored. The user can check later.

As described above, in this example, the user of the terminal can easily receive the voice mail service and the schedule management service from the central processing unit 1 without impairing the operability.

[0118] In addition, the terminal, for example, when A ₁ is configured as shown in Figure 2, by storing the speech uttered by the user of the terminal A ₁ in the storage unit 8 of the terminal A _1, further voice Input operability can be improved.

Specifically, when the terminal A ₁ is configured as shown in FIG. 2, normally, the switch S ₁ is open and the switch S ₂ is closed. Therefore, the central processing unit
(Server) When there is a voice input request from the side ₁ , the voice (word) uttered by the user of the terminal A1 is not stored in the storage unit 8, but is directly stored in the central processing unit (server) through the line. When the user wants to store the voice uttered by the user in his / her terminal A ₁ , the user only has to press the recording button 41 of the terminal (telephone) A ₁ . When the recording button 41 is pressed, the switch S ₁ is closed and the switch S ₂ is also closed, so that the voice uttered by the user is written into the storage unit 8, and the central processing unit (server) 1 is connected via the line 2. Sent to

When the sound stored in the storage unit 8 of the terminal A ₁ , that is, the recorded sound is reproduced and transmitted to the central processing unit 1, the user operates the reproduction button 42 of the terminal A _1.
You just have to press. When the playback button 42 is pressed, the switch S
₁ is opened, closing the switch S _2, at this time, recorded voice is read out from the memory unit 8 is sent to a central processing unit (server) 1 side via a line 2.

Such a function is used, for example, in a case where the voice of a predetermined word by the same user is input from the request input unit 11 of the terminal A ₁ a plurality of times in order to strictly determine the operation right of the terminal user. It is advantageous.

Further, the request target cannot be specified by only one input from the request input means of the terminal, the request management means cannot instruct the processing to the information processing means, and the request management means of the information processing apparatus is again operated. It is also advantageous when inquiring the request input means of the terminal to prompt the next input.

That is, a function of storing input voice in the request input means 11 of the terminal and transmitting the voice stored in the storage unit 8 to the central processing unit 1 as necessary is further provided. In this case, the terminal can transmit the voice data stored in the storage unit 8 when the input of the same word is again requested from the central processing unit 1, and the user can speak again. You can save time and effort. For example, in response to a question from the central processing unit (server), fixed words such as “Yes” and “No” are stored in the storage unit 8.
, It is not necessary to utter a fixed word such as "yes" or "no" many times in response to a question from the central processing unit (server), thereby significantly improving operability. be able to.

[0124] As described above, in the present invention, the central processing unit 1, a request from each terminal A ₁ to A _n centrally managed, after is made connection between the terminal and the central processing unit 1, the terminal side When a predetermined item is transmitted by voice from the central processing unit 1, the central processing unit 1 outputs information corresponding to the predetermined item input from the terminal to the terminal by voice. The apparatus 1 prompts the terminal to input the next item to be input by voice, and performs information exchange with the terminal by voice in an interactive manner. Times voice input is required (e.g., when a terminal requests a server to output a voice message of a predetermined message to a predetermined destination, voice information input from the terminal includes at least a destination and a destination. And two pieces of information are needed. Even case) as a central processing unit 1
From the terminal, a voice prompting the user of the terminal what to pronounce next is output, so that the user of the terminal responds by voice only with respect to the instructions given by the central processing unit 1. Therefore, even when information is exchanged by voice,
Operability can be significantly improved.

Further, in the present invention, the central processing unit 1 prompts the terminal to input the next item to be input by voice, and in response to this, a predetermined item is input from the terminal by voice. When the item is transmitted, a voice recognition dictionary corresponding to the predetermined item is selected, and the voice of the predetermined item input from the terminal is recognized using the selected voice recognition dictionary. The necessary vocabulary can be selected in one utterance to reduce the number of vocabulary words to be collated, preventing the accuracy of speech recognition from deteriorating. For example, the same utterance is repeatedly output from the central processing unit 1 due to incorrect recognition. Can be avoided.

In FIG. 1 and FIG. 2, the central processing unit 1 is realized by, for example, a personal computer or the like.
ROM that stores a control program for the CPU
Alternatively, a RAM or the like used as a work area of the CPU may be provided.

Here, the information processing means 22 shown in FIGS.
Voice recognition means 24, request management means 25, voice synthesis means 2
6 is realized by a CPU, and the storage means 21 can be realized by a RAM or the like.

The functions of the CPU as such information processing means 22, voice recognition means 24, request management means 25, voice synthesis means 26 and the like are, for example, software packages (specifically, information such as a CD-ROM or the like). recoding media)
Can be provided in the form of

In other words, the central processing unit 1 of the present invention
C for general-purpose computer systems such as personal computers
The present invention can also be implemented in an apparatus configuration in which a program recorded on an information recording medium such as a D-ROM is read and a predetermined process is executed by a microprocessor of the general-purpose computer system. In this case, the information processing means 2
2. A program for executing the processing of the voice recognition means 24, the request management means 25, and the voice synthesis means 26 (ie,
The program used in the hardware system is provided in a state recorded on a medium. The information recording medium on which the program or the like is recorded is not limited to a CD-ROM, but may be a ROM, a RAM, a flexible disk, a memory card, or the like. The program recorded on the medium is installed in a storage device incorporated in the hardware system, for example, a hard disk device, and executes this program to execute the information processing unit 22, the voice recognition unit 24, and the request management unit 2.
5. The voice synthesizing means 26 can be realized.

A program for realizing the information processing means 22, the voice recognition means 24, the request management means 25, and the voice synthesis means 26 of the present invention is provided not only in the form of a medium but also by communication (for example, (Provided by the server).

[0131]

As described above, according to the first to sixteenth aspects of the present invention, at least one terminal and an information processing function are provided, and information is exchanged with at least one terminal. And a central processing unit for performing information processing in the information processing system, the central processing unit centrally manages requests from each terminal, and after the terminal side and the central processing unit are connected, the terminal When a predetermined item is transmitted by voice from the side, the central processing unit is configured to output information corresponding to the predetermined item input from the terminal to the terminal by voice. The device prompts the terminal to input the next item to be input by voice, and information is exchanged with the terminal by voice in an interactive manner. Can be used, Et al, in between the central processing unit and the terminal, can not impair the operability, realizing various processing functions. More specifically, complex information processing can be carried out interactively between a highly portable terminal and a central processing unit by voice input / output, and the user of the terminal can perform operation by voice input / output. Various services (services such as voice mail and schedule management) can be received from the central processing unit (server) without impairing the performance. In addition, since input from the terminal can be performed by voice, the terminal does not require a special device and can use a normal telephone.

In particular, according to the second and fourteenth aspects of the present invention, the central processing unit confirms the operation right of the terminal user with respect to the terminal after the connection with the terminal. When the terminal prompts the user to input the necessary items by voice, and in response to the terminal sends a voice to confirm the operation right of the user of the terminal, the feature amount of the voice is registered in advance. The user's operation amount of the terminal is confirmed by checking the characteristic amount of the user of the terminal, and the use of the device of the user other than the user registered in advance can be prevented, and the information can be protected. You can do things.

Further, according to the twelfth and sixteenth aspects of the present invention, the central processing unit prompts the terminal to input the next item to be input next, and in response to this, the terminal transmits a predetermined item from the terminal. When the item is input by voice and the predetermined item is transmitted by voice, a dictionary for voice recognition corresponding to the predetermined item is selected, and the voice of the predetermined item input from the terminal is used by using the selected voice recognition dictionary. In the case of voice input, the required vocabulary can be selected in one utterance at the time of voice input, and the number of vocabularies to be collated can be reduced, preventing the accuracy of voice recognition from being reduced.
For example, it is possible to avoid a situation where the same utterance is repeatedly requested from the central processing unit 1 due to erroneous recognition.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration example of an information processing system according to the present invention.

FIG. 2 is a diagram showing a more specific configuration example of the terminal shown in FIG.

FIG. 3 is a diagram showing a more specific configuration example of the central processing unit shown in FIG. 1;

FIG. 4 is a diagram illustrating an example of a program structure of an information processing unit.

FIG. 5 is a diagram illustrating a configuration example of a voice recognition unit.

FIG. 6 is a diagram illustrating a configuration example of a recognition dictionary database.

FIG. 7 is a diagram illustrating a configuration example of a speech synthesis unit.

FIG. 8 is a diagram illustrating an example of a data structure of a start program of a request management unit.

FIG. 9 is a diagram illustrating an example of a data structure of an authentication program of a request management unit.

FIG. 10 is a diagram showing an example of a data structure of a menu program of the request management means.

FIG. 11 is a diagram showing an example of a data structure of a voice mail program of a request management unit.

FIG. 12 is a diagram illustrating an example of a data structure of a schedule management program of a request management unit.

FIG. 13 is a diagram showing an example of a data area secured in a storage unit of the central processing unit during a voice mail service.

FIG. 14 is a diagram illustrating an example of a data area secured in a storage unit of the central processing unit during a schedule management service.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Central processing unit 2 Line 3 Concentrator 4 Network 6 Telephone body 7 Handset 8 Storage unit 11 Request input unit 12 Information output unit 21 Storage unit 22 Information processing unit 23 Request reception unit 24 Voice recognition unit 26 Voice synthesis unit 27 Information transmission Means A _{1 to An} terminal S ₁ , S ₂ switch 41 Record button 42 Play button

Claims (16)

[Claims] 1. An information processing system comprising at least one terminal and a central processing unit having an information processing function and exchanging information with at least one terminal. Request input means for transmitting the input voice to the central processing unit when a request for the processing device is input as voice, and information output means for receiving and outputting information from the central processing device by voice ,
Further, the central processing unit includes: an information processing unit for performing predetermined information processing; a storage unit for storing data necessary for the information processing; a voice transmitted from a request input unit of the terminal; Voice recognition means for voice recognition of
Manages information recognized by the voice recognition unit, and when the voice information recognized by the voice recognition unit is a request to the information processing unit, performs a process corresponding to the request to the information processing unit. Request management means, and voice synthesis means for voice-synthesizing information to be transmitted from the central processing unit to the terminal, converting the information into voice, and transmitting the generated voice to the terminal. Information processing system. 2. The information processing system according to claim 1, wherein when the terminal is connected to the central processing unit, the request management means of the central processing unit specifies a user to the terminal. When there is a voice input for specifying the user from the request input means of the terminal in response to the request, a voice recognition means of the central processing unit is provided with a request input means of the terminal. By comparing the feature of the voice input from the terminal with the voice feature of the user of the terminal registered in advance, it is determined whether or not the user of the terminal has the operation right to the central processing unit. Characteristic information processing system. 3. The information processing system according to claim 2, wherein the request management means prompts the terminal to input a voice by the same user a plurality of times. When the voice is input a plurality of times, the voice recognition unit determines whether the user of the terminal has an operation right to the central processing unit based on a feature amount of the voice of the same user input a plurality of times. An information processing system characterized by determining. 4. The information processing system according to claim 1, wherein the request management means re-determines the target of the request based on a voice input from a request input means of the terminal. An information processing system for making an inquiry by voice to a request input unit of the terminal. 5. The information processing system according to claim 1, wherein said terminal has a function as a telephone. 6. The information processing system according to claim 1, wherein the request input means of the terminal has a storage unit for storing voice data uttered by the user, and the voice uttered by the user as required. Information processing means for storing data in a storage unit, and, if necessary, reading out audio data stored in the storage unit and transmitting the audio data to the central processing unit. system. 7. The information processing system according to claim 1, wherein the request management means of the central processing unit urges the terminal to input a voice about a mail, and in response to the request, inputs a voice request for the terminal. When there is voice input for mail from the input means, the information processing means of the central processing unit has a function of storing input voice data from the request input means of the terminal in a data storage area of another user of the storage means. An information processing system, comprising: 8. The information processing system according to claim 7, wherein the information processing means is configured to store the voice data from the predetermined user in a data storage area of another user of the storage means. An information processing system having a function of notifying the content of the voice data to the terminal of the other user by voice from the voice synthesis unit. 9. The information processing system according to claim 1, wherein the request management unit of the central processing unit prompts the terminal to input a voice about a user's schedule, and in response to the request, When there is a voice input about the schedule from the request input means of the terminal, the information processing means of the central processing unit, the content of the schedule voice input from the request input means of the terminal, the voice corresponding to the date and time, An information processing system having a function of storing data in the storage means. 10. The information processing system according to claim 9, wherein the information processing means uses the voice synthesizing means to store the contents, date, and time of the schedule stored as voice data in the storage means. An information processing system having a function of transmitting a voice to a computer. 11. The information processing system according to claim 9, wherein said information processing means stores a content of a schedule stored as audio data in said storage means in correspondence with a content of said schedule. And on time,
An information processing system having a function of notifying the terminal by voice from the voice synthesis unit. 12. The information processing system according to claim 1, wherein the request management means prompts a user of the terminal to input a predetermined target voice, and in response to the input voice being transmitted from the terminal. Selecting a dictionary for voice recognition corresponding to the target of the voice input from the terminal in order for the voice input from the terminal to be voice-recognized by voice recognition means, and providing the voice recognition means with the selected dictionary for voice recognition. An information processing system for recognizing an input voice from a terminal by using a personal computer. 13. An information management method in an information processing system, comprising: at least one terminal; and a central processing unit having an information processing function and exchanging information with at least one terminal, The central processing unit centrally manages requests from each terminal, and after a connection is made between the terminal and the central processing unit, when a predetermined item is transmitted by voice from the terminal side, the central processing unit receives an input from the terminal. The information corresponding to the given item is output to the terminal by voice, and at each stage of this processing, the central processing unit prompts the terminal to input the next item to be input by voice. An information management method characterized in that information is exchanged by voice with the user in an interactive manner. 14. The information management method according to claim 13, wherein the central processing unit is required to confirm the operation right of the terminal user to the terminal after the connection with the terminal. When the terminal prompts the user to input an important item by voice, and in response to the terminal transmits an item for confirming the operation right of the user of the terminal by voice, the feature amount of the voice and the pre-registered An information management method, comprising checking a feature amount of a terminal user and confirming an operation right of the terminal user. 15. The information management method according to claim 14, wherein the central processing unit performs a process of confirming an operation right of a terminal user, and when the central processing unit approves the operation right to the terminal user, the central processing unit notifies the terminal of the operation right. On the other hand, the input of a request to the central processing unit is prompted by voice, and in response to this, when a request is input by voice from the terminal, information corresponding to the request is transmitted to the terminal by voice. Information management method. 16. The information management method according to claim 13, wherein the central processing unit prompts the terminal to input a next item to be input by voice, and in response to the request, the terminal transmits a predetermined item from the terminal. When voice is input and the predetermined item is transmitted by voice, a dictionary for voice recognition corresponding to the predetermined item is selected, and the voice of the predetermined item input from the terminal is used by using the selected voice recognition dictionary. An information management method characterized by voice recognition.