Data processing: speech signal processing – linguistics – language – Speech signal processing – Application
Reexamination Certificate
1998-03-02
2003-03-04
Knepper, David D. (Department: 2654)
Data processing: speech signal processing, linguistics, language
Speech signal processing
Application
Reexamination Certificate
active
06529875
ABSTRACT:
TECHNICAL FIELD
This invention relates to a voice recognition device and voice recognition method for recognizing input voices, and a game machine using same.
BACKGROUND ART
With the progress of computer technology in recent years, video game machines using computer graphics technology have become widely used. A wide variety of different game software has been developed recently, so domestic video game machines of this kind have become extremely popular as a form of entertainment. In a video game machine of this kind, characters are moved and the game screen is changed by operating buttons on a controller.
In a conventional video game device, all game instructions are given by means of the player operating controller switches. However, for humans, the most natural means of communication is the voice. Therefore, various attempts have been made to recognize the human voice and to control machines on the basis of these recognition results For example, attempts have been made to control cars, robots, or the like by means of the human voice, or to control various electronic devices by vocal commands.
However, when it is sought to apply a conventional voice recognition device to a video game machine, the device must conform to the characteristics of the game machine, unlike the case of general electronic devices, and it must be adapted before application. For example, when the player presses a button on the game machine controller, rather than pressing the button just once, the degree of movement of the character can be changed by pressing the button for a long time or pressing it repeatedly, and the character can be made to perform special actions by pressing the button simultaneously with another button, for example. It is also important that the device has a good response. On the other hand, in some types of game, the accuracy of recognition may not be particularly important, indeed, the fact that the device cannot recognize the player's voice very well may itself increase the enjoyment of the game.
In brief, for the player to enjoy the game, it is necessary to provide particular characteristics and functions which differ from those in a conventional voice recognition device. A voice recognition device which satisfies these demands is not known in the prior art.
This invention was devised in order to resolve the related problems, an object thereof being to provide a voice recognition device and voice recognition method suited for a game.
It is a further object of this invention to provide a game machine using this voice recognition device and voice recognition method.
DISCLOSURE OF INVENTION
In a voice recognition device used as a peripheral device for a game machine, the voice recognition device relating to this invention comprises voice input means, and a voice recognition section for recognizing the player's voice by comparing the voice signal output from this voice input means with data from previously defined voice recognition dictionaries, and generating control signals relating to the game on the basis of the recognition result.
In the voice recognition device relating to this invention, the voice recognition section comprises a non-specific speaker voice recognition dictionary, which is previously defined for unspecified speakers, and a specific speaker voice recognition dictionary which is defined by the player, and in its initial state, the device selects the non-specific speaker voice recognition dictionary.
In the voice recognition device relating to this invention, the voice recognition section comprises a plurality of specific speaker voice recognition dictionaries corresponding respectively to a plurality of players, and one of these specific speaker voice recognition dictionaries is selected by an action of the player as the dictionary to be used for voice recognition processing.
The voice recognition device relating to this invention comprises a game machine control section connected to the voice recognition section, and the voice recognition section generates control signals relating to the game by combining voice recognition results and control signals from the control section.
In the voice recognition device relating to this invention, the control section outputs control signals for implementing normal actions, and the voice recognition section generates control signals for implementing special actions.
“Normal actions” are actions normally implemented by the player in the game (for example, kicking, jumping, etc.) “special actions” are particular actions which are made possible by the combination of signals from the control section and voice recognition signals. For example, in a fighting game, by a combination of a raised voice+operation of button A, the player may implement an action whereby the same move is repeated, or a particular deathblow which can be produced only when a plurality of buttons are pressed simultaneously.
In the voice recognition device relating to this invention, the voice recognition section outputs a value indicating the state of the voice signal and a similarity level indicating the degree of similarity between the voice signal output from the voice input means and the contents of the voice recognition dictionary, and the voice recognition section has a first operating mode, wherein voice recognition is conducted on the basis of the similarity level in order to select the type of game action, and a second operating mode, wherein no voice recognition is conducted and the state of the voice signal is measured in order to set the state of the action.
In the second mode, the voice recognition device relating to this invention takes the average sound level of a part or all of the voice signal as the state of the voice signal, and sets the state of the game action on the basis of this average sound level.
In the second mode, the voice recognition device relating to this invention takes the peak sound level of the voice signal as the state of the voice signal, and sets the state of the game action on the basis of this peak sound level.
In the second mode, the voice recognition device relating to this invention takes the voice signal rise time as the state of the voice signal, and sets the state of the game action on the basis of this rise time.
In the second mode, the voice recognition device relating to this invention takes the voice signal continuation time as the state of the voice signal, and sets the state of the game action on the basis of this continuation time.
In the second mode, the voice recognition device relating to this invention takes the type of voice as the state of the voice signal, and sets the state of the game action on the basis of this voice type.
In the voice recognition device relating to this invention, the voice recognition section outputs a similarity level indicating the degree of similarity between the voice signal output from the voice input means and the contents of the voice recognition dictionary, and the corresponding volume level, evaluates-this volume level on the basis of a predetermined rejection level, and rejects the recognition result from the voice recognition section depending on this evaluation result.
The voice recognition device relating to this invention sets a rejection level for each type of game or each game stage.
In the voice recognition device relating to this invention, the voice recognition section performs voice recognition on the basis of the similarity level indicating the degree of similarity between the voice signal output from the voice input means and the contents of the voice recognition dictionary, and changes the state of action in the game according to control signals generated on the basis of this recognition result, in response to this similarity level.
The game machine relating to this invention comprises the voice recognition device as a peripheral control device.
The voice recognition method relating to this invention comprises: a first step whereby a voice signal is received; a second step whereby the player's voice is recognized by comparing this
Hamada Kazuhiko
Nakajima Takuya
Keating & Bennett LLP
Knepper David D.
Sega Enterprises Ltd.
LandOfFree
Voice recognizer, voice recognizing method and game machine... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Voice recognizer, voice recognizing method and game machine..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Voice recognizer, voice recognizing method and game machine... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3049196