JP2002041279A - Agent message system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform indirect communication smoothly by bringing variety into a message without using an image of a message sender. SOLUTION: By use of a voice obtained through a microphone 11, a system generates and replays message data. In the system, there is provided (a1) a means 24 setting feeling of the message sender as a parameter by input operation, (a2) a database 16 storing agent image data having a plurality of modes coping with the feeling to be parameterized, (a3) an image selecting means 25 obtaining the agent image data coping with the set parameter, (a4) a message data organizing means 26 organizing the message data by use of the voice obtained through the microphone and the agent image data obtained by the image selecting means, (b1) a monitor 14, (b2) a speaker 12, and (b3) a means 27 replaying the message data in response to a requirement of a message receiver.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a message system using an agent image, and to a system configuration for realizing non-real-time communication.

[0002]

2. Description of the Related Art Conventionally, there are various types of message systems. The message system records and accumulates the voice of a message sender, and reproduces the accumulated message when a recipient of the message performs a predetermined operation.

As an example of such a system, there is a system using a telephone network. A user records and registers a message in a message center, which is a base station, using a telephone line. Then, the message recipient who has obtained the information indicating that the message message is registered in the message center can reproduce the message message by inquiring of the message center.

[0004] In this way, even when a message recipient cannot be contacted in real time, indirect communication through the message system is enabled. Also,
A so-called answering machine function of a telephone terminal is also one form of the message system.

[0005]

However, the above-mentioned message system records only voice information as a message, and has a problem that the message becomes inorganic. Further, in a message system using only voice, an actual intention of expression may not be accurately transmitted to a message recipient, and may cause misunderstanding.

[0006] Therefore, a message system in which a message sender is photographed by an imaging camera to create message data by voice and image is also conceivable.

However, since it is necessary to equip the terminal used by the message sender with an imaging camera, there is a problem that the cost of the terminal increases, and it is not preferable to use the imaging camera as an essential element of the configuration. In addition, since the configuration of the terminal is large, restrictions are particularly increased in a portable terminal and the like.

[0008] In some cases, it may be difficult to store an image of a person's figure as message data in the first place. Further, even in the case where content that is difficult to talk about or content that is difficult to rely on is left as a message message, it may be desirable to avoid saving an image of one's own figure as message data.

[0009] In view of the above problems, the present invention provides a system that allows message messages to be diversified without using an image of the message sender, thereby facilitating indirect communication. .

[0010]

According to a first aspect of the present invention, there is provided a message receiver which generates message data from a voice of a message sender obtained from a voice input means and stores the message data in a message database. In the system for reproducing the message data in response to the request, as a component related to the creation of the message data, (a1) by the input operation of the message sender, the parameter quantified the emotion of the message sender Parameter setting means to be set as, (a2) an agent image database storing agent image data having a plurality of modes corresponding to emotions to be parameterized, and (a3) corresponding to the parameters set by the parameter setting means. Image selection for acquiring agent image data from the agent image database And, (a4) message data organizing means for organizing the message data from the voice obtained by the voice input means and the agent image data obtained by the image selecting means, for reproducing the message data. As related components, (b1) image output means for displaying the agent image data, (b2) sound output means for reproducing the sound obtained by the sound input means, (b3) in response to a request of the message recipient Message data reproducing means for accessing the message database and reproducing the message data using the audio output means and the image output means.

According to a second aspect of the present invention, there is provided the agent message system according to the first aspect, further comprising: (a5) speech recognition of the speech acquired from the speech input means, as a component related to the creation of the message data. The parameter setting means comprises: (a1-1) means for estimating the message sender's emotion based on an analysis result of the voice analysis means and parameterizing the message sender. And

According to a third aspect of the present invention, in the agent message system according to the second aspect, the voice analysis means includes:
(a5-1) It is characterized by comprising means for analyzing the syntax of the voice acquired from the voice input means.

According to a fourth aspect of the present invention, there is provided the agent message system according to any one of the first to third aspects, further comprising: (a6) the message sender Environment detecting means for detecting the surrounding environment at the time of remarks, the parameter setting means further comprises (a1-2) means for parameterizing the environment information obtained by the environment detecting means. .

According to a fifth aspect of the present invention, in the agent message system according to the first or fourth aspect, the parameter setting means further comprises: (a1-3) an input operation of the message recipient. Means for changing the parameters of the message data being organized, wherein the message data organizing means comprises (a4-1) means for reorganizing the message data in accordance with the changed parameters.

According to a sixth aspect of the present invention, there is provided the agent message system according to any one of the first to fifth aspects, wherein components relating to creation of the message data and components relating to reproduction of the message data. Are realized by being provided in one terminal.

A seventh aspect of the present invention is the agent message system according to any one of the first to fourth aspects, wherein the agent message system includes a message creation terminal and a message reproduction terminal connected via a network. Wherein the message creation terminal includes a component related to creation of the message data and the message database, and the message playback terminal includes a component related to playback of the message data. .

The invention of claim 8 is the agent message system according to any one of claims 1 to 4, wherein the agent message system comprises a message creation terminal and a message reproduction terminal connected via a network. And the message server, wherein the message creation terminal includes components related to creation of the message data, the message server includes the message database, and the message playback terminal involves playback of the message data. It is characterized by having a component.

[0018]

Embodiments of the present invention will be described below with reference to the drawings.

{1. First Embodiment: Stand-Alone Type FIG. 1 is a configuration diagram of a stand-alone type system in which an agent message system of the present embodiment is realized in a single terminal 30. In this embodiment, the terminal 30 is a personal computer, but may be a stationary telephone terminal, a mobile telephone terminal, a message system dedicated terminal, or the like.

The terminal 30 has a microphone 1 as a voice input means.
1, a speaker 12 as an audio output unit, and a monitor 14 as an image output unit. The environment detection unit 15 is a unit that detects an external environment of the terminal 30, and is, for example, an illuminance sensor or a temperature sensor. The control unit 20 can input the external environment of the terminal 30 by these environment detecting means 15.

The operation unit 18 is a manual input operation unit including a pointing device such as a mouse, a keyboard, and the like. The user can execute various functions of the terminal 30 by using the operation unit 18. Setting input is enabled.

The control unit 20 provided in the terminal 30
A central part that controls the operation of the system, including the CPU 21 and the like, effective control of various means related to creation of message data described below, access control to each database, and execution control of various means related to reproduction of the message data. And so on.

As a component related to the creation of message data, the terminal 30 includes, as shown in FIG.
2. Parameter setting means 24, image selecting means 25, message data organizing means 26, agent image database 1
6 and the terminal 30 has a message database 17. In addition, as described later, when the parameter setting using the voice analysis is not performed, the voice analysis unit 22 is used.
Is not an essential component of the present embodiment. If the parameter setting using the surrounding environment of the message sender, which will be described later, is not performed, the environment detecting unit 15 is not an essential component of the present embodiment.

The parameter setting means 24 is a means for parameterizing the expression intention of the message sender who inputs the message message through the microphone 11, that is, the emotion at the time of speaking the message. The message sender
In parallel with the operation of inputting a voice to the microphone 11, an emotional state input when a message is being made is input. The input operation is performed by the parameter setting means 2 displayed on the monitor 14.
4 may be input using the operation unit 18 according to a user interface (not shown).

When the emotion of the message sender changes in the middle of the message, an input operation is performed using the changed emotion as a parameter in parallel with the input of the voice. As a result, a plurality of parameters that change over time are set in the message message.

The input items include, for example, “very good” and “good or bad” as a single parameter.
An ambiguous input such as “good”, “normal”, “bad”, “very bad” may be requested, or a numerical input such as “+5 to −5” directly quantified may be requested. You may make it.

As input items, for example, “degree of anger”, “degree of fun”, “degree of apology”,
A plurality of types of parameters such as "degree of appeal" can be input, and vague input or numerical value input can be similarly input to each parameter.

In this embodiment, in addition to the manual parameter setting described above, parameter setting by voice analysis is possible.

The voice analyzing means 22 is a means for recognizing and syntactically analyzing the voice inputted by the microphone 11. First, the voice input from the microphone 11 is converted into a text by being recognized by voice. by this,
Characteristic expressions can be extracted from the input speech. Next, by analyzing the syntax of the converted text, it is possible to extract more detailed characteristic expressions from the input speech. Note that the speech analysis unit 22 may be configured as a unit capable of executing only speech recognition, in addition to the configuration capable of executing both speech recognition and syntax analysis as in the present embodiment.

The parameter setting means 24 estimates the emotion of the person who inputs the voice (the message sender) based on the result of the analysis by the voice analysis means 22 described above, that is, based on the characteristic expression of the voice. Further, a function for setting ambiguous information of emotion as a quantified parameter is further provided.

For example, if an imperative expression is extracted from the speech recognized by the speech analysis means 22, it is determined that the message sender is angry, and the feeling of anger is parameterized. It is. If an expression indicating an apology is extracted by the voice analysis means 22, it is determined that the message sender has apologized, and the emotion of the apology is parameterized.

As described above, the terminal 30 is connected to the voice analysis unit 2.
By providing 2, the expression intention of the message sender can be automatically parameterized from the voice uttered by the message sender. The parameter may be a single parameter or a plurality of parameters, similarly to the parameter setting by the manual operation described above.

In this way, the voice of the message sender can be acquired by the microphone 11 and the expression intention of the message sender can be acquired as a parameter automatically or manually. However, the configuration may be such that the voice analysis unit 22 is not provided, and the parameter setting may be performed only by manual operation.

The agent image data is a substitute image for transmitting a message message on behalf of the message sender, and is stored in the agent image database 16. The agent image data may be, for example, an anime character or an animal character,
A fictional person may be created. Further, images of a plurality of characters and persons may be stored. When agent image data of a plurality of characters is stored, it is preferable that the message sender can select an appropriate agent image based on the message content and the relationship with the other party (the message receiver). Therefore, in the present embodiment, it is possible to select which agent image data is to be used as the substitute image in the operation unit 18.

Then, the agent image database 1
Reference numeral 6 has a plurality of types of agent image data corresponding to the parameters set by the parameter setting means 24. For example, the agent image data of the monster character includes a plurality of images corresponding to the degree of anger, a plurality of images corresponding to the degree of fun, and the like.

In other words, the agent image data, which is a substitute image, enables the expression intention of the message sender to be transmitted through the image. It is not always necessary to provide the agent image data in a one-to-one correspondence with the set values of the parameters, and the agent image data may be made to correspond to the set values in a certain range on a many-to-one basis.

The image selecting means 25 operates according to the parameters set by the parameter setting means 24 described above.
This is means for searching the agent image database 16 and acquiring corresponding agent image data.

Here, when the agent image data is set as a plurality of parameters which change with the passage of time as described above, a plurality of (aspects) agent image data are obtained corresponding to the parameters. Will be.

The message data organizing means 26 organizes, as message data, the agent image data obtained by the execution of each of the above-mentioned means and the voice data obtained by digitizing the voice inputted by the microphone 11.

If the agent image data is set as a plurality of parameters that change with the passage of time as described above, a plurality of (aspects) agent image data are acquired corresponding to the parameters. That is, the message data organizing means 24 organizes the plurality of agent image data as a moving image. That is, the message data is organized as message data composed of agent image data composed of moving images and audio data, and is a data file capable of synchronously reproducing moving images and audio. The composed message data is stored in the message database 17.

FIG. 2 shows a process for creating message data in the above configuration. First, a voice is input from the microphone 11. Then, parameters are set in the operation unit 18 or by performing voice analysis. The parameter setting by voice analysis may be performed together with the parameter setting by the operation unit 18.

Next, the image selecting means 25 acquires the agent image from the agent image database 16 based on the set parameters. Then, the message data compiling means 26 composes message data composed of the voice data and the agent image data.

When the message data is stored in the message database 17 as described above, the message data is reproduced in response to a request from the message receiver. That is, when the message receiver performs a key input or a mouse operation with the operation unit 18 and performs a message confirmation request operation, the message data is reproduced by the message data reproducing unit 27, and the voice of the message sender and the agent Image data is
These are output from the speaker 12 and the monitor 14, respectively.

Further, means for notifying the message recipient of the presence or absence of a message may be provided. For example, a message indicating that a message from a message sender is stored may be displayed on the monitor 14.

FIG. 3 shows an example of an implementation by the above-described agent message system. In the example shown in the figure, the message sender is a child, and a message message is recorded for the father. For example, if a child who has been mischievous during the day wants to apologize to his father, he / she leaves a message using terminal 30. At this time, the degree of apology may be set high in the operation unit 18 as a parameter,
If the voice input from the microphone 11 includes an expression indicating an apology such as "I'm sorry," the parameters are automatically set. At this time, the operation unit 18 selects “monster character” as the agent image data.

Then, the father who came home late at night
, The presence or absence of a message is confirmed, and if a message message remains, the message data reproducing means 27 is executed to reproduce the message data. by this,
The image of the monster character is displayed on the monitor 14 and the sound data is reproduced, and the monster character apologizes to the father instead of the child. The monster character is agent image data organized according to the parameter settings, and an image expressing an apology is displayed. If the parameters are set in accordance with the passage of time, the monster character is displayed as a moving image, and is dynamically displayed in a manner according to the apology message.

As described above, by using the agent image data, compared to the message system using only voice,
It is possible to accurately convey the expression intention of the message sender. In addition, the father can leave a reply message to the apology message of the child, and it is possible to indirectly communicate with the father who returns home late.

It is also possible to set parameters according to the environment around the message sender. As shown in FIG. 1, the terminal 30 includes an environment detecting unit 15. The environment detecting unit 15 is, for example, an illuminance sensor or a temperature sensor, and sets parameters according to the external environment detected by the environment detecting unit 15.

For example, when the environment around the message sender input from the illuminance sensor is a "bright" environment, the parameter setting means 24 makes a positive setting such as "bright" or "fun" as a parameter. What should I do? Also, if the environment around the message sender input from the temperature sensor is a "warm" environment,
Settings such as “cheerful mood” and “easy mood” may be made. However, it is preferable to freely set what parameters are to be set for what environment, and it is sufficient that the operation unit 18 can set rules for parameter setting.

The configuration described above is a stand-alone type configuration. As shown in FIG. 4, the terminal 30 for creating message data has a system configuration also serving as a terminal for reproducing message data. On the other hand, the peer-to-peer system configuration shown in FIG. 5 will now be described.

{2. Second Embodiment: Peer-to-Peer Type The peer-to-peer type agent message system shown in FIG. 5 includes a message creation terminal 31 for creating message data, and a message playback terminal 32 for playing message data. The creation terminal 31 and the message reproduction terminal 32 are connected via a communication network 40. Communication network 40
Is telephone line, ISDN line, Internet, LAN
It is not particularly limited.

FIG. 6 shows the configurations of the message creation terminal 31 and the message reproduction terminal 32, respectively. The message creation terminal 31 includes components related to creation of message data, that is, the microphone 1
1, a voice analyzing unit 22, a parameter setting unit 24, an image selecting unit 25, a message data organizing unit 26, an agent image database 16, and a message database 17. Note that, as in the first embodiment, when parameter setting by voice analysis is not performed, the voice analysis unit 22 is not an essential component.

On the other hand, the message reproduction terminal 32 includes components related to reproduction of message data, that is, the speaker 12, the monitor 14, and the message data reproduction means 27.

With such a configuration, similar to the first embodiment,
The message data is organized in the message creation terminal 31 and stored in the message database 17. Then, in the message reproducing terminal 32, when the message recipient executes the message data reproducing means 27, the message data reproducing means 27 accesses the message database of the message creating terminal 31 via the communication interfaces 42, 42, and transmits the message data. In addition to downloading, the message data is reproduced using the speaker 12 and the monitor 14.

In the present embodiment, the components relating to the creation and reproduction of message data are configured such that the message creation terminal 31 and the message playback terminal 32 share the functions, respectively. However, this is shown as a minimum configuration. The message data may be transmitted and received between the terminal 30 of the embodiment shown in FIG. 1, that is, the terminal including both components related to creation and reproduction of the message data. With such a configuration, message messages can be exchanged in both directions.

In the present embodiment, the message data is downloaded from the message creation terminal 31. However, the message creation terminal 31 may not create the message data, but may perform only the parameter setting. In this configuration, the message playback terminal 32 downloads only the parameters and retains the agent image database 16 in the message playback terminal 32 so that the message playback terminal 32 organizes and plays the message data. do it.

{3. Third Embodiment: Center Server Interposed Type Next, an embodiment of a center server type agent message system shown in FIG. 7 will be described. As shown in FIG. 7, in the present embodiment, a message creation terminal 33 that creates message data, a message server 34 including the message database 17, and a message playback terminal 35 that plays back the message data. 33, the message server 34,
The message playback terminal 35 is connected via a communication network 40.

FIG. 8 shows the configurations of the message creation terminal 33, the message server 34, and the message reproduction terminal 35, respectively. The message creation terminal 33 includes components related to creation of message data, that is, the microphone 11, the voice analysis unit 22, the parameter setting unit 24, the image selection unit 25, and the message data organization unit 2.
6. An agent image database 16 is provided.
Note that, similarly to the first and second embodiments, when the parameter is not set by voice analysis, the voice analysis unit 22 is not an essential component.

The message server 34 has a message database 17.

On the other hand, the message reproducing terminal 32 includes components related to the reproduction of message data, that is, the speaker 12, the monitor 14, and the message data reproducing means 27.

With such a configuration, similar to the first embodiment,
Although the message data is organized in the message creation terminal 31,
The message data organizing means 26 does not accumulate the organized message data locally (terminal 33), but accumulates it in the message database 34 of the message server 34 via the communication interfaces 43 and 44.

Then, in the message reproducing terminal 35, when the message receiver executes the message data reproducing means 27,
The message data reproducing means 27 includes a communication interface 45,
The user accesses the message database 17 of the message server 34 through 44, downloads the message data,
The message data is reproduced using the speaker 12 and the monitor 14.

In this embodiment, similarly to the second embodiment, the components related to the creation and reproduction of message data are
Although the message creation terminal 33 and the message playback terminal 35 are configured to share functions, this is shown as a minimum configuration, and the terminal 30 of the embodiment shown in FIG. Terminals having both components related to reproduction may be configured to transmit and receive message data via the message server 34. With such a configuration, message messages can be exchanged in both directions.

{4. Modification In the above-described embodiment, all the parameter setting units 24 reflect the information of the message sender. That is, parameters are set by manual input operation by the message sender, analysis of the voice of the message sender, and detection of the surrounding environment of the message sender. On the other hand, the parameter may be changed by reflecting the information of the message recipient (the person who plays the message message).

For example, in the first embodiment shown in FIG. 1, the parameter setting means 24 can change the parameters of the message data already organized. After confirming the presence or absence of the message message, the message receiver inputs the emotional state of the message receiver by the parameter setting means 24 prior to the reproduction of the message data.

For example, if the mood of the message recipient is very bad, a parameter with a high degree of “bad mood” is set. As a result, the parameters of the already composed message data are changed, the agent image data is acquired by the image selecting means 25 in response to this, and the message data is reorganized by the changed agent image data.

As described above, by reorganizing the message data by reflecting the emotion of the message receiver, the message receiver can control his / her own emotion. For example, when you are irritated, you can calm down by talking calmly, and when you feel dark, you can change your mind by talking brightly.

It should be noted that the parameter setting means 24 should be able to appropriately adjust what parameters to set when inputting the emotional state of the message recipient in accordance with the tastes of the user. .

As described above, by reflecting the emotions of the message receiver, there is an advantage that the message receiver can control his / her own emotion, and also the message sender can realize smooth communication. There is an advantage.

[0070]

As described above, according to the first aspect of the present invention, message data is organized from the message sender's voice obtained from the voice input means and the agent image data. The emotion is set based on the quantified parameters of the emotion set by the user's input operation, so that the message can be transmitted by voice and image, so that the communication can be reliably performed. Further, since the agent image is used instead of the image of the person, the present system can be used without any difficulty. Further, since the agent image is selected so as to accurately represent the message sender's expression intention, smooth communication is possible while being indirect.

According to the second aspect of the present invention, since the voice of the message sender obtained from the voice input means is recognized and the message sender's emotion is estimated from the result of the voice recognition and parameterized, the parameter setting is automatically performed. Operability is improved.

According to the third aspect of the present invention, in addition to the voice recognition means, a parsing means is further provided to estimate the message sender's emotion from the parsing result and parameterize the message sender. Can be transmitted more reliably. Further, since the parameter setting is performed automatically, the operability is excellent.

According to the fourth aspect of the present invention, the environment around the message sender is detected by the environment detecting means, and the environment information is parameterized, so that the message data can be adapted to the environment.

According to the fifth aspect of the present invention, the parameter setting can be changed by the input operation of the message receiver, and the message data is rearranged according to the changed parameter. Can be organized. This also facilitates communication.

In the invention according to claim 6, 7, or 8, the agent message system can be used in various system forms. The system configuration according to claim 6 is suitable for a case where a plurality of people (family, colleagues, etc.) exchange message messages at the same place, such as at home or in the same department, and the like. The system configuration 8 is suitable for exchanging message messages between remote persons.

[Brief description of the drawings]

FIG. 1 is a diagram showing a standalone system configuration of an agent message system.

FIG. 2 is a diagram showing a process of generating and reproducing message data by an agent message system.

FIG. 3 is a diagram showing an example of an implementation of an agent message system.

FIG. 4 is a schematic diagram of a stand-alone system.

FIG. 5 is a schematic diagram of a peer-to-peer system.

FIG. 6 is a configuration diagram of a peer-to-peer type system.

FIG. 7 is a schematic diagram of a center server type system.

FIG. 8 is a configuration diagram of a center server type system.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 Microphone 12 Speaker 14 Monitor 16 Agent image database 17 Message database 22 Voice analysis means 24 Parameter setting means 25 Image selection means 26 Message data organization means 27 Message data reproduction means

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G10L 13/00 H04M 3/42 J 5K024 21/06 3/533 5K101 15/10 11/10 15/00 G10L 3/00 S 19/00 531N H04M 3/42 551A 3/533 9/18 J 11/10 F-term (reference) 5B049 CC02 DD01 DD03 DD05 EE01 EE07 FF03 FF04 FF06 FF09 GG04 GG07 5D015 AA01 AA05 HH00 HH21 KK02 LL12 5D45 DB01 5E501 AB18 AC15 BA15 BA17 CA03 CA08 CB15 DA13 EA21 FA14 FA32 FB25 5K015 AB02 AD05 GA02 5K024 AA72 BB05 CC02 EE09 FF04 FF06 5K101 KK02 MM07 NN07 NN15 NN18 NN21 NN36 NN37

Claims (8)

[Claims] 1. A system for creating message data from a voice of a message sender obtained from a voice input means, storing the message data in a message database, and reproducing the message data in response to a request from a message receiver. (A1) parameter setting means for setting the message sender's emotion as a quantified parameter by the input operation of the message sender, and (a2) corresponding to the emotion to be parameterized. (A3) an image selecting means for acquiring agent image data corresponding to the parameters set by the parameter setting means from the agent image database; ) The voice acquired by the voice input means and the image selection And message data organizing means for organizing the message data from the agent image data obtained by the selecting means. (B1) Image output means for displaying the agent image data as components relating to reproduction of the message data (B2) voice output means for reproducing the voice obtained by the voice input means, (b3) in response to a request of the message recipient, access the message database, the message data the voice output means and And a message data reproducing means for reproducing using the image output means. 2. The agent message system according to claim 1, further comprising: (a5) a voice analysis unit that performs voice recognition of a voice acquired from the voice input unit, as a component related to creation of the message data. An agent characterized by comprising: (a1-1) means for estimating an emotion of the message sender based on an analysis result of the voice analysis means and parameterizing the message sender. Message system. 3. The agent message system according to claim 2, wherein the voice analyzing unit includes: (a5-1) a unit that performs a syntax analysis of a voice acquired from the voice input unit. Agent messaging system. 4. The agent message system according to claim 1, further comprising: (a6) surroundings at the time of the message sender's speech as a component related to creation of the message data. An agent message system, comprising: (a1-2) means for parameterizing the environment information obtained by the environment detecting means. . 5. The agent message system according to claim 1, wherein said parameter setting means further comprises: (a1-3) a message already organized by an input operation of said message receiver. Means for changing data parameters; and the message data organizing means comprises: (a4-1) means for reorganizing the message data in accordance with the changed parameters. 6. The agent message system according to claim 1, wherein a component related to creation of the message data and a component related to reproduction of the message data are one unit. Agent message system that is realized by the terminal. 7. The agent message system according to claim 1, wherein the agent message system includes a message creation terminal and a message reproduction terminal connected via a network, An agent message, wherein the message creation terminal includes a component related to creation of the message data, and the message database, and the message playback terminal includes a component related to playback of the message data. system. 8. The agent message system according to claim 1, wherein the agent message system includes a message creation terminal, a message reproduction terminal, and a message server connected via a network. The message creation terminal comprises: a component related to creation of the message data; the message server comprises: the message database; and the message playback terminal: a component related to playback of the message data; An agent message system comprising: