JP2001314649A - Voice game method and apparatus, and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform a communication with a user in a different manner by a method wherein a common electronic character distinguishes a plurality of users. SOLUTION: A voiceprint analyzing part 6 is provided in a control part 1 to analyze the voiceprint of a user to identify the user. The part 6 is different from a voice recognizing part 7 which recognizes the meaning of a specific voice but distinguishes the voice which is being generated by a user from any voices. An algorithm is incorporated into the control part 1 to vary response, judgment, and reply of the electronic character corresponding to a distinguished user.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an audio game method, apparatus, and recording medium, and more particularly, to an audio game method, apparatus, and recording medium in which the correspondence of electronic characters changes according to the user.

[0002]

2. Description of the Related Art Conventionally, as a voice recognition game device for recognizing voice, a device which converts a human voice into a digital signal, identifies the digital signal as a player's intention in a game machine body, and reflects it in a game is known. I have. More specifically, it is composed of a game machine main body, a game cartridge in which voice game software is incorporated, and a voice recognition system including a microphone.

[0003] The game cartridge is mounted on the main body of the game machine as in a normal game. On the other hand, the microphone is connected to a controller terminal of a vacant game machine body via the voice recognition system. When the power is turned on, an electronic character appears on a television screen connected to the game machine body. When the user speaks to the electronic character in the screen, the voice is recognized by the voice recognition system via the microphone. The recognition result enters the game machine body. The game machine body analyzes and analyzes the voice, converts the voice into a word that can be interpreted by the game machine body, and transmits it to the electronic character. The electronic character responds or acts in response.

On the other hand, although it is not for recognizing voice, what is called character-type mailer software is also known. Specifically, on the Internet and e-mail,
Letting an electronic character as a pet carry mail,
It takes care of the electronic character on the screen.

[0005]

However, the above-mentioned prior art has the following problems. (1) Voice Recognition Gaming Device When a user calls, an electronic character reacts,
It is possible to give the user a sense of unity of playing with the electronic character. However, since the speech recognition accuracy is low, the user is limited to a specific speaker and cannot be expanded to non-specific speakers. Therefore, the same electronic character could not be shared by a plurality of users. In addition, the number of recognizable words is about 10 words, which has many restrictions, and there is a limit to the spread of the game while being a language input game.

(2) Character-type mailer software An electronic character whose main job is to carry mail can be loved as a pet, so that it is possible to enjoy mail operations that tend to be tasteless and dry. However, since the button operation is mainly performed, there is a sense of incompatibility compared with the real pet, and since the interactive conversation cannot be performed, there is an unnatural and unsatisfactory aspect. In addition, since this is also assumed to be a pet dedicated to an individual, a plurality of users cannot share the same electronic character as in the above-described voice recognition game device (1).

An object of the present invention is to solve the above-mentioned problem of the prior art by allowing a plurality of users to recognize a specific user, and to provide a user with a correspondence of an electronic character common to a plurality of users. It is an object of the present invention to provide a voice game method, a device, and a recording medium that can be changed in accordance with each other and can provide a sufficient sense of satisfaction.

[0008]

According to a first aspect of the present invention, there is provided a voice game method in which the correspondence of the electronic character changes according to the owner of the voice uttered toward the electronic character displayed on the display unit. The electronic character correspondence refers to the behavior and facial expression of the electronic character, the conversation content spoken by the electronic character, the manner of speaking, and the like. A common electronic character can identify a plurality of users and perform different correspondence communication for each user. As a result, the game becomes more realistic.

According to a second aspect of the present invention, the features of the voiceprint of the user are registered in advance as a user, the voiceprint of the voice uttered toward the electronic character is analyzed, and the features of the voiceprint are extracted. 2. The voice print feature registered by the user, and when both the features match, the owner of the voice is specified, and the behavior and correspondence of the electronic character are changed according to the specified owner. It is a voice game method described. By a simple method of analyzing the voiceprint of the user's voice, it is possible to specify the user who could not be recognized by voice recognition.

According to a third aspect of the present invention, there is provided a voice input unit for inputting a voice uttered by a user, a voiceprint analysis unit for analyzing a voiceprint of the voice input from the voice input unit and extracting characteristics of the voiceprint, By comparing the features of the voiceprint extracted by the voiceprint analysis unit with the features of the voiceprint registered in advance by the user,
When both features match, specifying means for specifying the user who uttered the voice, and control means for changing the correspondence to the electronic character according to the user specified by the specifying means,
A display unit for displaying the electronic character in response to a display command from the control means.

According to a fourth aspect of the present invention, there is provided a process for extracting a feature of a voiceprint from a voice uttered by a user, a process for identifying a user by comparing the extracted feature of a voiceprint with a feature of a registered voiceprint. A computer-readable recording medium storing a program for causing a computer to execute a process of changing the correspondence of the electronic character according to the user and a process of displaying the electronic character according to a display command.

[0012]

Embodiments of the present invention will be described below. FIG. 1 is a block diagram of a voice game device according to an embodiment. The audio game device includes a control unit 1, a display unit 9,
It mainly includes a voice input unit 5, a voice determination unit 8, an operation unit 2, an external storage device 3, a voice synthesis unit 10, and a voice output unit 11.

The control unit 1 is a game machine main body in which a storage device, a control device, an arithmetic device and the like are incorporated, and controls the peripheral devices connected thereto. The display unit 9 is configured by a television monitor connected to the control unit 1 and displays an electronic character or the like serving as a common pet for a plurality of users. The voice input unit 5 collects voice uttered by the user.
For example, it is constituted by a microphone. The voice judgment unit 8
A voice recognition unit 7 for analyzing the voice collected by the voice input unit 5
And a voiceprint analyzer 6 for analyzing a voiceprint of a voice. The voice recognizing unit 7 determines a predetermined 1
The voice of the keyword of about 0 words is analyzed, the meaning of the keyword is recognized, and the recognition result is applied to the control unit 1 and converted into a digital signal that the electronic character can understand. The voiceprint analysis unit 6 analyzes the voiceprint of the voice uttered by the user, extracts its features, specifies one user from among a plurality of users, adds the specified result to the control unit 1, and converts the result into a digital signal.

The operating means 2 is composed of, for example, a controller. When the control unit 1 of this embodiment is applied to a personal computer, an input device such as a keyboard and a mouse is used as the operation unit 2. External storage device 3
Stores image data and program data for image processing and breeding processing of games and the like, algorithms and keywords for specifying users, and also stores audio data such as music and sound effects as necessary. The stored data is read into the control unit 1 as needed. Usually, it is composed of a ROM cartridge.
A CD-ROM, a magnetic disk, or the like may be used instead of the cartridge. The voice synthesizer 10 forms a predetermined voice signal by performing voice synthesis according to a command from the controller 1.
The voice output unit 11 outputs a voice or the like of the electronic character based on the voice signal added from the voice synthesis unit 10. For example, it is composed of a speaker.

An electronic character is displayed on the display unit 9.
The electronic character performs different responses, judgments, replies, and the like in accordance with a voice uttered by the user according to a predetermined algorithm stored in the external storage device 3. In addition, interactive training with the user enables different training depending on the user.

Here, the operation of the voice judging section 8 will be described. The voiceprint analysis unit 6 registers a user's voiceprint features and frequency components in association with a user name that is the main voiceprint of the user. On the other hand, an algorithm capable of recognizing a specific keyword is incorporated in the voice recognition unit 7 in advance. The reason why keywords are limited to specific ones is that accurate speech recognition is very difficult, and therefore, the keywords are limited to those having a high hit rate.

When the user utters an arbitrary voice or a specific keyword toward the voice input unit 5, the voice is applied to the voiceprint analysis unit 6 on the one hand. Here, voiceprint analysis is performed, and the owner of the voice is specified. The algorithm compares the characteristic part and frequency component of the user's voiceprint registered in advance with the characteristic part and frequency component of the voiceprint of the input voice, and compares the voiceprint of the input voice with the registered voiceprint. When a match is found, the owner of the matched voiceprint is selected from the user registration data to identify the owner of the voiceprint. This algorithm is commonly used when registering a voiceprint in advance.

On the other hand, the speech is also applied to a speech recognition unit 7. Here, voice analysis is performed. If the voice is the specific keyword, a digital signal corresponding to the meaning of the keyword is input to the control unit 1. The voiceprint analysis unit 6 mainly contributes to identification, and the voice recognition unit 7 mainly contributes to training.

Here, the role of performing voiceprint analysis in addition to speech recognition will be described. Since the method of raising the electronic character differs depending on the user, if the user changes, the degree of growth and the growth direction of the electronic character naturally differ. By specifying a user by voiceprint analysis, a storage area for training data and the like relating to an electronic character that differs for each user is prepared for each user. Further, in addition to the method of raising the user, an algorithm for making the correspondence of the electronic character different for each specified user from the beginning is created. Speech recognition is common to all users,
No matter who the user says a particular keyword, they are all interpreted the same.

It should be noted that the above-described speech judgment unit 8 and speech synthesis unit 10 have been described as peripheral devices connected to the control unit 1.
It is also possible to incorporate it in the control unit 1.

Next, the operation of the audio game apparatus according to the embodiment will be described with reference to FIG. When the user turns on the power switch 4 of the apparatus, the control section 1 causes the display section 9 to display an electronic character (step 201). A predetermined voice signal synthesized by the voice synthesis unit 10 in response to a command from the control unit 1 is added to the voice output unit 11 and uttered "You who".
At this time, the mouth of the electronic character is moved so that the electronic character projected on the screen is uttering toward the user (step 203). After the utterance of the electronic character, the user utters himself or herself in response to the utterance. The control unit 1 determines the presence or absence of the voice from the user via the voice input unit 5 and the voice determination unit 8 (step 205). When the voice from the user is detected, the voiceprint analysis unit 6 determines that the voice has the owner. Someone is judged by comparing with the user registration data (step 207). When the user who has made the voice can be specified, the control unit 1 checks whether or not the specified user has a message (step 209).

If there is a message, the electronic character is displayed as "○
○, there is a message "(step 211).
"OO" indicates a user name. For example, if the user is the owner of the house and is registered as "Dad", the message is "Dad, there is a message." If the specified user is the patriarch, the honorific may be used, such as "Dad, there is a message." After the utterance of the electronic character, the user is asked whether or not to reproduce the message (step 213). When the reproduction is requested, the message is reproduced and transmitted to the user (step 215). Thereafter, the process proceeds to a menu screen display step 225 described later. If the user cannot be specified in step 207 for specifying the user, the electronic character utters "I do not know who" (step 217), and displays the user registration list on the display unit 9 (step 219). The display contents are, for example, as follows.

Registration number User name 1. Dad 2. Mom 3. Ichiro:: If the user's name is included in this list, it is not registered with the user name "OO" that was issued earlier,
It means that it was registered under another name. For example, the user utters his or her nickname "Onii", but is registered in the list with his or her name "Ichiro". In this case, if the user does not change the registered name, the user specifies the user by uttering the number “3” of the registered list. Conversely, if an attempt is made to re-register the user name or if the user name is not included in this list, a new registration is performed. This branch determination is made in step 221. When a registered user is selected from the list, it is determined that the user has been specified, and the process proceeds to step 209 for determining whether there is a message. If a registered user is not selected, the process proceeds to a user registration routine (step 223) to perform user registration.

As shown in FIG. 3, in the above-described user registration routine, the user waits for the voice of the user to be registered (step 301), and if there is, the voice is recorded (step 303), and the voiceprint analysis is performed. The characteristic and the frequency component are detected by the unit 6 (steps 305, 3
07), user registration is performed so that the user can be specified from the voice of the user (step 309).

When the user registration is completed, a menu screen is displayed on the display unit 9 (step 225), and the electronic character utters "what?" (Step 2).
27) The user is prompted to select a mode from the menu. An example of the menu is as follows.

Menu Screen Number Mode 1. Message 2. Game 3. Quiz 4. Email 5. If the user utters the number of the desired mode, or utters the mode name, or clicks a predetermined portion with a mouse (step 229), the mode shifts to each selected mode. . Hereinafter, each mode will be described.

1. Message mode (FIG. 4) The electronic character utters "Who?"
1). The control unit 1 waits for the user to say the message partner (step 403). When the user name "OO" of the message partner is uttered, the control unit 1 determines who the message partner is from the registered user name (step 405). When the electronic character knows who the message is, the electronic character uses an expression to match the message, for example, "say!" (Step 407) to urge the user to speak the message. Here, the voice of the user is recorded for a predetermined time (step 40).
9). This recording may be performed by an automatic voice recording start function. After the recording is completed, the process returns to the step 225 for displaying the menu screen.

When it is not possible to confirm who the message partner is in step 405, the electronic character utters "I do not know who it is!" (Step 411). After the utterance, the registration list of the message partner is displayed on the display unit 9 (step 41).
3). The display contents are, for example, as follows.

Message Registration List Number User Name Dad 2. Mom 3. Ichiro:: If the user name is included in this list, it means that the user name "OO" which was generated earlier was not registered but was registered under another name. For example, the user uttered “Onii” meaning his brother, but was registered in the list with the name “Ichiro”. In this case, if the registered name is not changed, the registration list number "3" is uttered by voice and selected from the list (step 415). After the utterance, the process proceeds to the above-mentioned “say!” Step 407 to continue the message mode.

If the user does not make a selection and does not emit a voice, the user is asked whether to newly register (step 41).
7). If you do not register, the message mode will end,
It returns to the menu screen display step 225. If a new registration is to be made, the process proceeds to a message partner registration routine (step 419).

As shown in FIG. 5, when a voice to be registered is uttered (step 501), the voice is recorded (step 503) and the voice is analyzed by the voice recognition unit 7 (step 501). Then, the user name is registered in the message destination registration list (step 506). After registration,
The process proceeds to step 407 of “say” to continue the message mode.

2. Game Mode (FIG. 6) The display of the electronic character is displayed on the display unit 9 (601). It is determined whether there is a voice (step 60).
3) The user faces the electronic character, for example, "Hey!" When you say something or say something, the control unit 1
Analyzes the voice, converts it into a signal that can be understood by the electronic character, and transmits the signal to the electronic character (step 605). Thus, the electronic character performs a reaction, a reply, and an action according to the user's speech (step 60).
7). The reaction, reply, and operation content are different for each user. For example, the greater the number of times of speaking, the more intimacy appears in the expression of the electronic character. Thereby, game characteristics can be further improved. The communication between the user and the electronic character is repeated until the talking ends (step 6).
05-609). When the talking is completed, the process returns to the menu display screen step 225.

3. Quiz mode The quiz in this mode may be a publicly known one such as "riddle riddle" or "guess at a sightseeing spot". For these, it is conceivable to perform a question with a different degree of difficulty based on the past win or loss of the registered user. . Specific examples are omitted.

4. Mail Mode (FIG. 7) The display of the electronic character is displayed on the display unit 9 and the basic operation list screen is displayed in the margin of the display unit 9 (step 701). The basic operation list screen includes, for example, a basic operation list number and an operation item. E-mail 2. Check mail:: Items such as: are listed.

The user selects an operation number (step 7)
03). Here, it is assumed that "1. send mail" is selected. Then, the electronic character hears "Who is it?" In the manner of listening to the user (step 705).

It is determined whether or not there is a voice from the user (step 707), and when the user says the name of the person sending the mail to the electronic character "OO", the control unit 1 determines who the "OO" is. It is determined whether or not it can be determined (step 709). For example, if the electronic character is identified by collating it with the data in the registered friend book, the electronic character utters the voice mail content “say!” To the user (step 711). When the user continues to speak the contents of the mail toward the electronic character by voice, the control unit 1 records the voice as it is (step 71).
3). After the recording is completed, the control unit 1 transmits a voice mail to the specified mail account destination (step 715).
After the transmission is completed, the process returns to the menu screen display step 225.

In the step 709 for specifying the person sending the mail, when the name issued by the user is not known by the voice recognition unit 7, the electronic character sends a message "I do not know who!" (Step 71
7). Then, the mail destination registration list is displayed on the display unit 9 (step 719). A display example of the mail destination registration list is as follows.

Email address registration list No. Friends 1. Kohei 2. Haruo 3. Nobuchan:: If the name of the person sending the e-mail that the user said is included in this list, the name of the person who originated earlier, "○
"O" means that it was not registered but registered under a different name. For example, the user uttered the name “Nobuyasu”, but was registered in the list with the nickname “Nobuchan”. In this case, if the registered name is not changed, the registration list number "3" is uttered by voice and selected from the list (step 721).
After utterance, "say" mentioned above! To step 711, and continue the mail mode.

If the user does not select any item from the list and does not emit a voice, the user is asked whether or not to newly register (step 72).
3). If no new registration is made, the mail mode ends and the process returns to the menu display screen step 225. When newly registering, a mail registration routine (step 725)
Proceed to.

As shown in FIG. 8, in the mail registration routine, when a voice to be registered is issued (step 8).
01), the voice is recorded (step 803), analyzed by the voiceprint recognition unit 8, and the user name and other necessary data such as a mail address are registered (step 807). Note that the mail address may be input from the operation means 2 such as a controller. After registering, say
, And the mail mode is continued as it is.

5. Exit without doing anything If you select number 5 in the list on the menu screen, the game ends and the power switch 4 is automatically turned off.

As described above, since a user can be specified, a common electronic character can be kept, and since a user can be identified, data unique to the electronic character can be retained for each user.

As described above, according to the embodiment, since an algorithm capable of analyzing a user's voiceprint and discriminating a plurality of users is incorporated in a computer, it is possible for one person to perform voice recognition of a specific speaker lacking accuracy. As compared with the conventional voice game, various correspondences can be made to the same electronic character depending on the user. For this reason, a game can be enjoyed in a group or a family unit, and a new development of the game can be expected.

Although the above embodiment has been described with reference to a case where there are a plurality of users, the present invention can be applied to a case where there is only one user. For example, if a voiceprint at the time of health is registered and a cold voice at the time of catching a cold is registered, a different game can be developed even in a single game.

In the above embodiment, the breeding of the electronic character is not described in detail, but it can be bred as is known by the analysis by the voice recognition device. Further, a device for implementing the present invention is not limited to a game machine, but may be a personal computer.

In the embodiment, the registration lists are individually prepared, such as user registration, message destination registration, and mail destination registration, but these may be integrated into one. Further, a user registration routine, a message destination registration routine,
Although the mail registration routine is prepared separately, a subroutine may be used in common.

[0047]

According to the present invention, the correspondence of the electronic character is changed according to the owner of the sound emitted toward the electronic character displayed on the display unit. Compared with the game, the range of the play can be extended to the group, and the playability can be further enhanced.

[Brief description of the drawings]

FIG. 1 is a block diagram of an audio game device according to an embodiment.

FIG. 2 is a main flowchart according to the embodiment;

FIG. 3 is a flowchart of user registration according to the embodiment.

FIG. 4 is a flowchart of a message mode according to the embodiment.

FIG. 5 is a flowchart of registration of a message partner according to the embodiment.

FIG. 6 is a flowchart of a game mode according to the embodiment.

FIG. 7 is a flowchart of a mail mode according to the embodiment.

FIG. 8 is a flowchart of registration of a mail destination partner according to the embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Computer 2 Operation means 3 External storage device 4 Power switch 5 Voice input part 6 Voiceprint analysis part 7 Voice recognition part 8 Voice determination part 9 Display part 10 Voice synthesis part 11 Voice output part

Claims (4)

[Claims] 1. A voice game method in which the correspondence of an electronic character changes according to the owner of a sound emitted toward the electronic character displayed on a display unit. 2. Registering a user's voiceprint feature in advance as a user, analyzing a voiceprint of a voice uttered toward the electronic character, extracting the voiceprint feature, and extracting the extracted voiceprint feature and the user-registered voiceprint. The voice game method according to claim 1, wherein a feature is compared, and when the two characteristics match, the owner of the voice is specified, and the correspondence of the electronic character is changed according to the specified owner. 3. A voice input unit for inputting a voice uttered by a user, a voiceprint analysis unit for analyzing a voiceprint of the voice input from the voice input unit and extracting characteristics of the voiceprint, and a voiceprint analysis unit. Comparing the extracted voiceprint features with the voiceprint features registered in advance by the user, and when both the features match, specifying means for specifying the user who uttered the voice; and the user specified by the specifying means. An audio game device comprising: a control unit that changes the correspondence of the electronic character according to the control unit; and a display unit that displays the electronic character according to a display command from the control unit. 4. A process for extracting features of a voiceprint from a voice uttered by a user, a process for comparing a feature of the extracted voiceprint with a feature of a registered voiceprint to identify a user, Processing for changing the correspondence of the electronic character, processing for displaying the electronic character by a display command,
And a computer-readable recording medium for recording a program for causing a computer to execute the program.