JPH08187368A - Game device, input device, voice selector, voice recognizing device and voice reacting device - Google Patents

Abstract

PURPOSE: To provide a game device to be operated by a voice natural for a human being. CONSTITUTION: A voice recognizing part 2 recognizes a voice, a uttering zone detecting part 4 detects the uttering zone of a speaker (the operator of the device) from the motion near the lips of the speaker fetched by an image input part 3. Based on the voice recognized result and the information of the detected uttering zone, an integrative judging part 5 extracts only the voice recognizing result uttered by the speaker. The recognized result is sent to a control part 6 and utilized for controlling an airship 7. Thus, wrong recognition caused by surrounding noises other than the voice of the speaker can be prevented.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a game device operated by using a voice, an input device for inputting a lip image and a voice, and a voice reaction device.

[0002]

2. Description of the Related Art FIG. 34 shows an example of a conventional game machine.
1 shows a game device in which an airship equipped with a wireless receiver is operated by a remote controller with a wireless receiver in the hand of an operator. As shown in FIG. 34, in a conventional game device, a joystick 161 included in a remote controller is generally used to operate an object.
When the operator moves the joystick 161, the angle is detected by the angle detection units 162 and 163, converted into an electric signal and output to the control unit 164. The control unit 164 outputs a radio control signal for controlling the movement of the airship 7 according to the angle of the joystick 161 based on these electric signals.

[0003]

However, since the conventional game device is operated by the joystick 161, it is not a natural operation for humans. For this reason, there are problems that it takes a long time to master the operation, and that the reaction to the reaction becomes dull. Also, there is a game device that operates a balloon with a drive device instead of an airship, but in this device as well, since the movement of the balloon is controlled as described above, the movement of the balloon becomes abiotic, There was a problem that the warmth peculiar to balloons diminished.

An apparatus for recognizing a voice by inputting an image of an operator's lips has also been proposed. However, such an apparatus requires a sophisticated optical system lens, so that the apparatus itself is large. There is a problem that it is expensive and expensive.

The present invention has been made in view of the above circumstances, and the purpose thereof is (1) operation by natural voice is possible, operation familiarity is not required, and noise or voice is emitted. To provide a game device with a low cost and a simple structure that enables the use in a difficult situation and the use of a person with a speech disability. (2) Inputting the movement and voice of the operator's lips with a simple structure. An input device capable of performing, (3) a voice selection device that outputs as a voice a word randomly selected from a plurality of words with respect to the same input voice, (EN) A game device or toy that can be played, and a voice recognition device used therefor, and (5) a voice reaction device that can change its operation according to an input voice. There to be provided.

[0006]

The game device of the present invention comprises:
A voice input means for inputting at least one voice including a voice generated by an operator, converting the input voice into a first electric signal, and outputting the first electric signal, and the voice input means. Voice recognition means for recognizing the at least one voice based on the first electrical signal output from
Image input means for optically detecting the movement of the operator's lip, converting the detected movement of the lip into a second electric signal, and outputting the second electric signal, and the second electric signal. A generated section detecting means for receiving a signal and obtaining a section in which the voice is generated by the speaker based on the received second electric signal; and the at least one voice recognized by the voice recognition means. An integrated judgment means for extracting the sound generated by the operator from the at least one sound based on the section obtained by the generation section detection means, and the sound extracted by the integrated judgment means And a control means for controlling the object based on
This achieves the above object.

The vocalization section detecting means is a differentiating means for detecting the degree of change of the second electric signal output from the image inputting means, and the degree of change detected by the differentiating means is a predetermined value. And a means for determining that the corresponding voice is generated by the operator when the number exceeds the limit.

The integrated judgment means creates an evaluation section by adding a section of a predetermined length to the section obtained by the vocal section detection section, and the at least one recognized by the voice recognition section. One voice compares the recognition result output time with the means for detecting the recognition result output time outputted from the voice recognition means, and the recognition result output time of the at least one voice is compared. It may be provided with a unit for determining a voice contained in the evaluation section as the voice uttered by the operator.

Another game device of the present invention is an image input means for optically inputting the movement of the operator's lips, converting the inputted movement of the lips into an electric signal, and outputting the electric signal. A lip recognition unit that obtains the movement of the lip based on the electric signal, recognizes a word corresponding to the obtained movement of the lip, and outputs a recognition result, and a target according to a control signal based on the recognition result. And a control means for controlling the object,
This achieves the above object.

The lip recognition means selects one from the predetermined number of words in accordance with the storage means for storing a predetermined number of words and the obtained movement of the lip, and selects the selected word. A matching unit that determines that the word corresponds to the movement of the lip may be provided.

The storage means stores the movement of the lips corresponding to the predetermined number of words as a standard pattern, and the matching means stores all of the standard patterns.
It is also possible to calculate a distance from the obtained lip movement and select a word corresponding to one of the standard patterns having the smallest distance.

The game device inputs voice, converts the voice into another electric signal, and outputs the other electric signal, and the other electric signal output from the voice input unit. The control signal to be given to the control means based on both the voice recognition means for recognizing the voice based on the above, the recognition result by the voice recognition means, and the recognition result by the lip recognition means. It may further include integrated determination means.

The game apparatus includes means for obtaining a voice recognition reliability for the recognition result by the voice recognition means, and means for obtaining a lip recognition reliability for the recognition result by the lip recognition means. The integrated determination means selects one of the recognition result by the voice recognition means and the recognition result by the lip recognition means based on the voice recognition reliability and the lip recognition reliability. , It may be output as the control signal.

The image input means includes a light emitting means for emitting light and a light receiving means for receiving the light reflected by the lips of the operator and converting the received light into the second electric signal. You may have.

The image input means has a light emitting means for emitting light and a light receiving means for receiving the light reflected by the lips of the operator and converting the received light into the electric signal. May be.

The image input means has a light emitting means for emitting light and a light receiving means for receiving the light reflected by the lips of the operator and converting the received light into the electric signal. May be.

The light may be applied to the lips from the side.

The light may be applied to the lips from the front.

The voice input means may have at least one microphone. The voice input means may include at least one microphone, and the at least one microphone and the light emitting means and the light receiving means of the image input means may be provided on one stand.

The input device of the present invention comprises a headset in the form of a headphone, a support having one end joined to the headset, and a base joined to the other end of the support. A pedestal provided with at least one light-emitting element that emits light applied to the lips of a person and at least one light-receiving element that receives the light reflected by the lips. Achieve the purpose.

A voice input means for inputting voice may be provided on the table.

The voice selection device of the present invention is the first storage means for storing a plurality of tables, each of the plurality of tables including a plurality of words that can be output with respect to one input. A first storage means; a second storage means for storing one of the plurality of tables; and the first storage means stored in the second storage means in response to an external input.
Selecting means for selecting one word from the plurality of words included in one table and outputting the selected one word as voice, and the one table stored in the second storage means Of the plurality of tables stored in the first storage means to another table that is determined according to the selected one word, To achieve the above objectives.

The voice selection device may further include means for generating a random number, and the selection means may use the random number to select the one word from the plurality of words.

Another voice selection device of the present invention is a storage means for storing a table, and the table includes a storage means containing a plurality of words that can be output in response to one input, and external storage. Selection means for receiving an input, selecting one word from the plurality of words contained in the table stored in the storage means by using a random number, and outputting the selected word as a voice, and generating the random number And means,
This achieves the above object.

The voice reaction device of the present invention comprises the above-mentioned voice selection device and voice recognition means for inputting a voice, recognizing the voice, and giving a recognition result to the voice selection device.
This achieves the above object.

Another game device of the present invention is equipped with the above-described voice reaction device, thereby achieving the above object.

Another game device of the present invention comprises a plurality of the above-mentioned voice reaction devices, whereby the voice reaction devices interact with each other, thereby achieving the above object. Another game device of the present invention is a plurality of voice input units for converting input voices into electric signals, the plurality of voice input units respectively corresponding to different directions, and the electric signals. Energy of each of the plurality of voice input units is determined, one of the plurality of voice input units having the largest energy is determined, and the direction corresponding to the determined one voice input unit is determined. And a direction detecting means for determining that the direction is the direction in which the voice is generated, thereby achieving the above object.

The game apparatus may further include operation means for moving the object, and control means for controlling the operation means so as to change the operation direction of the object in the determined direction. Good.

The game device inputs measuring means for measuring the current direction of movement of an object and the determined direction, and obtains a target direction based on the current direction and the determined direction, The method further comprises direction selecting means having a means for storing the target direction and operation means for moving the object, the direction selecting means determining the difference between the target direction and the current direction. It may be used to control the motion means such that the current direction of motion of the object and the target direction substantially coincide.

In another game device of the present invention, input means for inputting a relative direction by voice, measuring means for measuring the current direction of the object, the current direction and the input relative direction are used. A game device provided with a direction selecting means having a means for obtaining a target direction based on a direction and storing the target direction, wherein the direction selecting means uses the difference between the target direction and the current direction. The object is controlled so that the current direction of the object and the target direction substantially coincide with each other, thereby achieving the above object.

The input means may include an input section for inputting the voice and a recognition section for recognizing the relative direction based on the input voice.

Another game device according to the present invention comprises input means for inputting an absolute direction by voice, means for determining a target direction based on the absolute direction, and storing the target direction, and an object. A direction selection means having a measurement means for measuring the current direction of the target device, wherein the direction selection means uses the difference between the target direction and the current direction to detect the current direction of the object. The object is controlled so that the direction of and the target direction substantially coincide with each other, thereby achieving the above object.

The input means may include an input section for inputting the voice and a recognition section for recognizing the absolute direction based on the input voice.

The voice recognition device of the present invention receives the electric signal corresponding to the voice, and detects the voice end point, which is the time when the input of the voice ends, from the electric signal; Second detection means for determining a vocalization section, which is a section in which the voice is uttered, of a section in which the voice is input, based on the signal, and a feature based on a part of the vocalization section of the electric signal. A feature quantity extraction means for creating a quantity vector, a storage means for storing the feature quantity vectors of a plurality of candidate voices created in advance, and the feature quantity vector from the feature quantity extraction means are stored in the storage means. Means for recognizing the input voice by comparing each of the feature amount vectors of the plurality of candidate voices,
This achieves the above object.

The first detecting means outputs the electric signal,
Means for dividing into a plurality of frames each having a predetermined length, calculating means for obtaining the energy of the electric signal for each of the plurality of frames, and determining the voice end point based on the dispersion of the energy It may be provided with a determining means for performing.

The deciding means decides the voice end point by comparing a predetermined threshold value with the variance of the energy, and the voice end point is determined by comparing the variance of the energy with the threshold value. May be the time when the variance matches the threshold when changing from a large value to a small value.

The determining means may use the variance of the energies of a predetermined number of frames among the energies of the plurality of frames.

The second detecting means includes means for smoothing the energy of the electric signal, and first circulating storage means for sequentially storing the energy of the electric signal without smoothing it for each frame. Second circular storage means for sequentially storing the smoothed energy for each frame, and the smoothed storage stored in the first cyclic storage means when the voice end point is detected. Threshold calculation means for calculating a vocalization section detection threshold value using both the non-energy and the smoothed energy stored in the second circulation type storage means, and the non-smoothed energy for the vocalization. It may have a vocalization section determining means for determining the vocalization section by comparing with a threshold for section detection.

The threshold value calculation means detects the maximum value of the unsmoothed energy stored in the first circulation type storage means at the time when the voice end point is detected, and the voice end point. The voicing section detection threshold value may be calculated by using the minimum value of the smoothing energy stored in the second circulation type storage means at a time when the vocalization section is not performed.

The feature amount detecting means detects the number of zero crossings for each frame of the electric signal and the number of zero crossings for each frame of the signal obtained by differentiating the electric signal from the vocalization section of the electric signal. And the energy of the electric signal are calculated, and these may be used as elements of the feature amount vector.

Another voice reaction device of the present invention comprises at least one voice recognition device described above, and at least one control means for controlling an object based on the recognition result of the at least one voice recognition device. Therefore, the above object is achieved.

The voice reaction device is connected to the at least one voice recognition device, and is connected to a transmission means for transmitting the recognition result by the at least one voice recognition device and to the at least one control device. Cage,
Receiving means for receiving the transmitted recognition result and giving it to the at least one control device, the at least one control device and the receiving means being attached to the object, whereby the object It may be possible to remotely control the object.

[0043]

In the game device of the present invention, the voice recognition means recognizes the input voice, and the utterance section detection device is the utterance section in which the speaker is uttering from the movement of the lip of the speaker (operator). To detect. Based on the voice recognition result and the voiced section detection result, the integrated determination unit recognizes the command input by the speaker by voice, and the control unit controls the object according to the command. As a result, it is possible to operate the game with a human voice, and it is possible to prevent an erroneous operation caused by erroneously recognizing a voice of a person other than the speaker. Further, in the other game device of the present invention, the command is directly recognized from the movement of the lips of the operator, so that it is possible to operate the game in the presence of noise by the human voice or in a situation where it is difficult to produce the voice. . In addition, this game device can be used by persons with disabilities. In still another game device of the present invention, the integrated determination unit determines a more probable recognition result from both the recognition result by the voice recognition means and the recognition result based on the movement of the lips. Therefore, in addition to the above-mentioned advantages, there is an advantage that the reliability of the game operation by voice can be further increased.

In the input device of the present invention, since a pillar is attached to a light headset, and an inexpensive light emitting element (eg, LED) and an inexpensive light receiving element (photodiode, etc.) are attached to a base attached to the pillar. It is possible to provide an input device that is extremely light and inexpensive. Furthermore, if the headset is made expandable and contractible, the length of the headset can be adjusted for each operator of the input device to adjust the positional relationship between the light emitting element and the light receiving element and the vicinity of the operator's lip. it can.

In the voice selection device of the present invention, when there is one input from the outside, one of the words contained in the table stored in the second storage means is selected and output as a voice. . Then, the table stored in the second storage means is changed to a table selected according to this output from among the plurality of tables stored in the first storage means. Next, when there is an input from the outside, the above operation is repeated. In this way, the voice selection device of the present invention can return not only one operation of returning one word to one input, but also returning words according to input given one after another. If this voice selection device is combined with a voice recognition device, it will recognize the corresponding words from the input voice, and depending on the recognition result,
A voice reaction device that outputs randomly selected words as voice can be configured. If at least one voice reaction device is provided in the game device, the voice reaction device can interact with the operator. If more than one voice reaction device is provided, a game device in which the devices interact with each other can be configured. it can. Further, by selecting a word to be output for one input by using a random number, it is possible to perform a different output instead of always outputting the same word for the same input.

In another game device according to the present invention, the direction in which a voice is input is detected by using a plurality of voice input units, each of which corresponds to a different direction. Then, the direction of movement of the object or the direction of the object itself is changed to the detected direction. In this way, the object can be operated by the voice. Further, in another game device of the present invention, the moving direction or the direction of the target object is changed while detecting the difference between the direction input by voice and the current moving direction or the direction of the target object.

The voice recognition device of the present invention detects the point at which voice input is completed from the electrical signal corresponding to the input voice. Subsequently, the section in which the voice is further uttered is extracted from the electric signal for the section in which the voice thus obtained is input. Since a feature amount vector that is actually compared with the feature amount vector of the candidate voice is created from the electrical signal for the section in which this voice is uttered, the voice recognition device of the present invention recognizes voice with high accuracy with a simple configuration. can do. Further, the threshold value used for extracting the section in which the voice is uttered is calculated based on the energy of the electric signal and a smoothed version of this energy.
As a result, it is possible to satisfactorily detect the section in which the voice is uttered. Furthermore, in the voice reaction device obtained by combining the voice recognition device with the means for controlling the motion of the target object, the target object can be caused to perform the motion corresponding to the input voice.

[0048]

【Example】

(First Embodiment) A first embodiment of the game apparatus of the present invention will be described below with reference to the drawings. The present embodiment is a game device for operating an airship with a voice command according to the movement of the airship. Voice commands are "front", "back",
It contains 6 commands, "Right", "Left", "Up" and "Down".

In this embodiment, a voice signal of the speaker (operator of the game device) and a signal representing the movement of the lip of the speaker are input, and whether or not the speaker is uttering based on these signals is input. The determination process is being performed. As a result, it is possible to prevent malfunctions due to ambient noise, especially voices spoken by others.

FIG. 1 briefly shows the structure of the game apparatus of this embodiment. The game device of this embodiment includes a voice input unit 1 and a voice recognition unit 2 for processing input voices, an image input unit 3 for inputting lip movements, and an image input unit 3 for processing signals indicating lip movements. The utterance section detection unit 4 is provided.
The voice recognition unit 2 and the utterance section detection unit 4 are both connected to the integrated determination unit 5, and determine what the command uttered by the speaker is based on both the voice input and the movement of the lips. To be done. The determination result of the integrated determination unit 5 is input to the control unit 6, and based on this, the control unit 6 causes the airship 7 to operate.
Control.

First, the voice including the command uttered by the speaker is input to the voice input unit 1. For voice input, for example,
A normal microphone or the like can be used. The voice input unit 1 converts the input voice into an electric signal and outputs it as a voice signal 11 to the voice recognition unit 2. The voice recognition unit 2 analyzes the voice signal 11 and outputs the result as a voice recognition result 12. The analysis of the audio signal 11 is performed, for example, by
It can be performed by a conventionally known method such as P matching.

In parallel with the above processing of the input voice, the processing of the electric signal representing the movement of the lips is performed. When the speaker utters a command, the movement of the lips at that time is input to the image input unit 3. FIG. 2 shows a configuration example of the image input unit 3. The image input unit 3 of the present embodiment irradiates the lip portion of the speaker with the light emitted from the LED 21, and the photodiode 22 detects the light reflected by the lip portion. As a result, the electric signal 13 corresponding to the movement of the lips is output. When the speaker's lip moves, the level of the electric signal 13 changes in accordance with the change in the shadow near the speaker's lip. The speaker's lips may be illuminated with light from the LED 21 from the front,
You may irradiate from a side surface.

The electric signal 13 from the image input section 3 is input to the vocal section detecting section 4. FIG. 3 shows the configuration of the vocalization section detection unit 4 of this embodiment. The utterance section detection unit 4 has a differentiating circuit 31 and a section detection unit 32. Differentiating circuit 31
Outputs a differential signal 33 indicating the degree of change of the input electric signal 13. An example of the waveform of the differential signal 33 is shown in FIG. FIG. 5 shows the differential signal 33 obtained when the speaker utters the commands “front” and “back” in a state in which the speaker's lips are laterally illuminated with the light from the LED 21. As can be seen from FIG. 5, when the speaker is speaking, the amplitude of the differential signal 33 is large. Further, since the LED light is applied to the speaker's lips from the side, it can be seen that the movement of sharpening the lips when the command "back""u" is emitted is reflected in the waveform. Note that when the light from the LED 21 is applied to the speaker's lip from the front, the light hits only the speaker's face, so that the electrical signal 13 and the differential signal 33 are not affected by noise caused by the movement of the background. There are advantages.

The section detection unit 32 receives the differential signal 33, determines the magnitude of the amplitude of the differential signal 33, and detects the speaking section of the speaker. FIG. 6 shows a specific method of detecting the vocalization section.
Will be described with reference to.

When the level of the differential signal 33 exceeds a predetermined amplitude threshold value 51, the section detecting unit 32 determines that the differential signal 33 is caused by the speaker uttering a command, and the differential signal 33 The section in which the level of exceeds the amplitude threshold 51 is the vocal section. In the example shown in FIG. 6, section 1 and section 2 are vocalization sections. Then, the intervals of the adjacent vocal sections are compared with a predetermined time threshold value 52. The time threshold value 52 is a value used for determining whether or not a plurality of vocalization sections correspond to the same vocalization, that is, whether or not a plurality of vocalization sections are continuous. If the interval of the vocalization section is within the time threshold 52, it is determined that the two vocalization sections sandwiching the interval are continuous vocalization sections. A signal 14 representing the continuous utterance section thus determined is output from the utterance section detector 4. The amplitude threshold 51 and the time threshold 52
Can be set to appropriate values in advance.

As described above, the utterance section detection unit 4
Detects the intensity and duration of lip movement using the differential signal 33 to obtain the section in which the speaker uttered a command.

Next, the operation of the integrated judgment section 5 will be described. As shown in FIG. 4, the integrated judgment unit 5 has a voice recognition time judgment unit 41, an output judgment unit 42, and an output gate 43. The voice recognition time determination unit 41 determines the voice recognition result 12
Is received and the time when the recognized voice is input to the voice input unit 1 is transmitted to the output determination unit 42. In addition to the output from the voice recognition time determination unit 41, the output determination unit 42 receives the utterance section detection signal 14 from the utterance section detection unit 4.
Here, the operation of the output determination unit 42 will be described with reference to FIG. 7.

The output judging section 42 first creates an evaluation vocal section 72 by adding a time threshold 71 for evaluation before and after the vocal section based on the received vocal section detection signal 14. Next, it is determined whether the time when the voice recognition result 12 is output from the voice recognition unit 2 is within the evaluation vocalization section 72. When it is within the range, it is determined that the voice input to the voice input unit 1 and recognized by the voice recognition unit 2 is uttered by the speaker. The result of the determination is output to the control unit 6 as the signal 15.

The time threshold value 71 for creating the evaluation utterance section 72 is set in consideration of the time required for the recognition process performed by the voice recognition unit 2. This is because the time when the voice recognition result 12 is output is used as one of the materials for determining whether the recognized voice is due to the utterance of the speaker.

In this way, when the signal 15 corresponding to the command input by voice is obtained, the control unit 6
Controls the airship 7 by outputting a radio control signal according to the input command.

As described above, in the first embodiment, the utterance section in which the speaker is uttering is detected from the movement of the lips when the speaker utters a command, and the recognized voice is detected based on this. Determine if it belongs to the speaker. Therefore, it is possible to prevent the erroneous recognition due to the utterance of a person other than the speaker, and the erroneous operation of the target object resulting therefrom.

Therefore, it is possible to realize a game device by a human-friendly operation of voice operation. In addition, in this embodiment, the movement of the lip of the speaker is LE
It is detected by a simple configuration / method such as a combination of D and a photodiode. Therefore, the image of the lip of the speaker can be realized at a very low cost as compared with the conventional device which has taken in the image using the video camera or the like.
Of course, a phototransistor may be used instead of the photodiode.

The circuit configurations shown in FIGS. 2 and 3 are merely examples, and the present invention is not limited to this configuration. It is also possible to realize it by using computer software.

(Second Embodiment) In the game device according to the second embodiment of the present invention, the command is not inputted by voice, but only by the movement of the lips, and the airship is moved according to the inputted command. Control. This enables use in a noisy environment, use in a situation where no voice can be produced, such as midnight, or use by a person with a voice disorder.

FIG. 8 is a schematic diagram showing the structure of the game apparatus of this embodiment. The game device according to the present embodiment includes the image input unit 3, the control unit 6, and the airship 7 as in the above-described first embodiment, and further, the lip for recognizing the words of the speaker (operator) from the movement of the lip. The recognition unit 81 is provided.

FIG. 9 shows an example of the structure of the lip recognition section 81. In the present embodiment, the lip recognition unit 81 includes a differentiating circuit 31, a difference calculation unit 91, a database 92, and a pattern matching unit 93. The differentiating circuit 31 is the same as that used in the utterance section detecting unit 4 of the game device of the first embodiment. The difference calculation unit 91 uses the differentiating circuit 31.
The differential signal 33 from is sampled in a predetermined time width,
Calculate the difference between the sampling data. The result of the difference calculation is sent from the difference calculation unit 91 to both the database 92 and the pattern matching unit 93. The database 92 holds the difference calculation results of standard patterns used for recognition. The pattern matching unit 93 obtains a difference in distance between the difference result of the held standard pattern and the difference calculation result of the input pattern that is the recognition target, and based on this difference, the word input as the movement of the lips Recognize. Of course, the smaller the difference, the higher the reliability of the recognition result.

The operation of the game apparatus of this embodiment will be described in detail below. In the present embodiment, the lip recognition unit 81 recognizes the standard pattern in advance by performing the recognition operation in order to recognize the input word by comparing the standard pattern with the input pattern as described above. It is necessary to register in 81.

(Registration Operation) First, the image input section 3 receives the LED reflected light reflected by the lip portion of the speaker and outputs the electric signal 13 corresponding to the movement of the lip to the lip recognition section 81. The electric signal 13 is input to the differentiation circuit 31 of the lip recognition unit 81. The differentiating circuit 31 transmits the differential signal 33 indicating the degree of change of the electric signal 13 to the difference calculating unit 91. Up to this point, the process is the same as in the first embodiment.

The operation of the difference calculator 91 will be described with reference to FIG. First, the differential signal 33 is sampled in a time width (Δt), and the difference between adjacent sampling data in the obtained sampling data is calculated. The difference between the calculated sampling data, that is, a series of difference data is output to the database 92. Database 9
2 holds this difference data string. The above operation is repeated for the number of words (categories) to be recognized, and difference data strings are stored for all categories. The stored difference data string is held as a standard pattern used for recognition. In this embodiment, the commands used to control the object are "front", "back", and "right".
・ The six items are "left", "top", and "bottom". Therefore, the above-described storage of the difference data string is repeated 6 times, and finally the 6 standard patterns are held in the database 92.

When all the standard patterns have been registered in the database 92 in this way, the database 92 examines each differential data string and determines the length of the section in which the data corresponding to the part where the lips are moving continues. Extract to the difference data string. Specifically, for example, if a value close to zero in the difference data string continues longer than a predetermined time, it is determined that the section corresponds to when the lips are not moving.
Then, when the length of the section corresponding to the part where the lips are moving is extracted for all the standard patterns, the standard pattern having the longest length is selected, and the length thereof is set to the difference data string length (N ). The registration operation is completed as described above, and the difference data string of the standard pattern is held in the database 92.

(During recognition operation) The operation from inputting the movement of the lip portion to obtaining the differential signal 33 is exactly the same as the registration operation. Here, the differential signal 33 is the difference calculation unit 9
The operation after the input to 1 will be described with reference to FIG.

The differential signal 33 input to the difference calculator 91
Is sampled in the time width (Δt) as in the registration operation. Next, the difference between adjacent sampling data is calculated for the sampling data within the section of the difference data string length (N) of the standard pattern, and the obtained series of difference data is used as the difference data string for the section. And The section in which the difference is calculated is sequentially shifted backward by Δt. In FIG. 11, the first sampling data is the beginning of the section, and the difference data string for the section 111 in which the section length is N and the section 111
Only the difference data string is illustrated for the section 112 that is shifted backward by N / 2 from 1 to.

When the difference data strings of a plurality of sections having the section length N (hereinafter, these are referred to as recognition difference data strings) are obtained, these recognition difference data strings are sent to the pattern matching unit 93. . Pattern matching unit 93
Reads the standard pattern from the database 92 and obtains the distance from each of the standard patterns for each of the plurality of recognized difference data strings. In this embodiment,
As described above, the six standard patterns are stored in the database 92.
Therefore, the pattern matching unit 93 calculates one distance from each standard pattern for each recognized difference data string.

The distance between the recognized difference data string and the standard pattern is calculated using the following formula. Here, r _i is the i-th recognized difference data string, p _ij is the j-th standard pattern (corresponding to the j-th category), and d _j is the distance between the recognized difference data string and the j-th standard pattern. When the distance d _j becomes a certain value or less, the pattern matching unit 93 determines that the recognized difference data string matches the jth standard pattern, and determines the signal 82 corresponding to the jth category (word). Output as a result.

The result of this determination is input to the control unit 6, and the control unit 6 outputs a radio control signal corresponding to the jth category to control the airship 7.

As described above, in this embodiment, the input word (command) is recognized based on only the movement of the lips,
Control the airship according to the recognized words. For this reason,
It is possible to use in a noisy environment, in situations where it is difficult to speak, and for people with speech disabilities.

The image input section 3 for inputting the movement of the lips
Can be realized by a combination of the LED 21 and the photodiode 22 as in the case of the first embodiment. Therefore, as compared with the conventional method of capturing the image of the lips itself using a video camera or the like, a very inexpensive game device can be provided. Can be provided.

In this embodiment, the user of the game registers a standard pattern used for command recognition prior to command input. However, for example, when manufacturing or shipping the game device, a standard pattern capable of coping with unspecified user's lip movements may be registered in the database 92 in advance, and registration by the user may be omitted. .

(Third Embodiment) Next, a game device according to a third embodiment of the present invention will be described. In the present embodiment, the airship is operated by inputting commands by both voice and movement of the lips of the speaker (operator) and integrating and recognizing the recognition results of both. Therefore, it is possible to reliably recognize the command uttered by the speaker even in a noisy environment.

FIG. 12 briefly shows the structure of the game apparatus of this embodiment. The game device according to the present embodiment includes a voice input unit 1, an image input unit 3, a control unit 6, and an airship 7 having the same configuration as that of the game device according to the first embodiment. Further, the audio processing unit 121 and the lip processing unit 122 are further provided. The voice processing unit 121 is the voice recognition unit 2 of the first embodiment.
Similarly, the input voice is recognized, and then the reliability of the recognition result is calculated. Further, the lip processing unit 122 recognizes a word (command) input as the movement of the lip similarly to the lip recognition unit 81 of the second embodiment, and calculates the reliability of the recognition result together with it. Outputs from the voice processing unit 121 and the lip processing unit 122 are both input to the integrated determination unit 123. The integrated determination unit 123 comprehensively determines the command input by the speaker based on the recognition results from the processing units 121 and 122 and the reliability, and outputs the determination result.

The operation of the game apparatus of this embodiment will be described in detail below.

Similar to the first embodiment, the voice input unit 1 inputs the voice uttered by the speaker (the operator of the game device) and transmits the electric signal 11 corresponding to the input voice to the voice processing unit 121. Is. The voice processing unit 121 receives the electrical signal 11 and recognizes the input voice based on the electrical signal 11.
As a voice recognition method, any conventionally known method may be used. Here, for example, similar to the method described in the description of the lip recognition unit in the above-described embodiment, data obtained by processing the electric signal 11 obtained when uttering all commands that may be input. The columns are registered in advance as a standard pattern, and the recognition target data sequence obtained by processing the electric signal 11 obtained when the operator of the game device actually speaks a command, and all of the previously registered data are registered. The command (voice) input from the voice input unit is recognized by calculating the distance from the standard pattern. When the voice is recognized in this way, the voice processing unit 121 subsequently obtains the reliability indicating how reliable the recognition result is, and integrates both the voice recognition result and the reliability as an output 124. Judgment unit 123
Give to. How to obtain the reliability will be described later.

Further, in parallel with the processing of the input voice, the processing of the signal representing the movement of the lips is performed. First, the image input unit 3 inputs the movement of the speaker's lip in the same manner as in Example 1, and the level of the electric signal 13 changes according to the movement of the lip.
Is output. The lip processing unit 122 receives the electric signal 13 and performs the same processing as in the second embodiment. However, the lip processing unit 122 of the present embodiment determines that the recognition difference data string is j as a result of the pattern matching between the recognition difference data string and the standard pattern.
If it is determined that the pattern matches the standard pattern of the th th, the distance d between the recognized difference data string and the standard pattern of the j th
The reliability of the recognition result is calculated based on _j . The recognition result and the reliability thus obtained are both the integrated determination unit 1
23 is output.

Next, the method of calculating the reliability will be briefly described. In this embodiment, the reliability of the voice recognition result and the reliability of the recognition result based on the movement of the lip are obtained by the same process. The calculation of the reliability of the voice recognition result will be described below.
Consider a case where the reliability of the voice recognition result is evaluated in three levels of “large”, “medium”, and “small”. The reliability of the recognition result is highest when the reliability is “small”, and the reliability of the recognition result is lowest when the reliability is “high”. In this case, the threshold α _L that separates the reliability “small” and “medium”, and the threshold α _H that separates the reliability “medium” and “large” (where α _L <α _H )
The distance d between the standard pattern determined to match the recognition target and the recognition target is compared with the threshold value. If the comparison result d <α _L , the reliability is determined to be “small”. Similarly, when α _L ≦ d <α _H and d ≧ α _H , the reliability is determined to be “medium” and “large”, respectively. Similarly, with respect to the recognition result based on the movement of the lips, the level of the reliability is determined by comparison with the threshold value. The threshold used here can be set to an appropriate value. Also,
The reliability calculation method is not limited to the method described here,
Any known method may be used.

Next, the operation of the integrated judgment unit 123 will be described with reference to FIG.
This will be described with reference to FIG.

FIG. 13 is a diagram showing the concept of a method for making an integrated judgment. First, the integrated determination unit 123 outputs the time when the voice recognition result is output from the voice processing unit 121 (that is, the time when the output 124 is generated) and the time when the recognition result based on the lip movement is output from the lip processing unit 122 ( That is, the time when the output 125 is generated is detected, and the sections corresponding to the predetermined threshold value 131 are added before and after each detected output time to create the evaluation sections 132a and 132b. Then, it is determined whether the evaluation section 132a for the lip recognition result and the evaluation recognition section 132b created for the voice recognition result overlap. If they overlap, the integrated determination unit 123 determines that the voice uttered by the operator who has input the movement of the lips has been input and has been recognized. If they do not overlap, it is determined that the recognized voice is due to noise in the surroundings or the utterance of something other than the operator. This can prevent erroneous recognition of voices other than the operator.

Next, the integrated judgment section 123 judges whether or not the recognition result based on the movement of the lips and the recognition result based on the voice match, and if they match, the recognition results are regarded as the integrated judgment result. (The integrated judgment result “before” in FIG. 13). If they do not match, the integrated judgment result is determined according to the reliability obtained for each recognition result. FIG. 1 shows an example of the correspondence relationship between the combination of the reliability of the recognition result and the integrated determination result determined according to the combination.
4 shows. In this example, as described above, the reliability with respect to each recognition result is evaluated in three stages of "most large", least reliable, "small", and "medium" between them. ing. FIG. 14A shows a correspondence relationship when the voice recognition result is prioritized when the reliability is the same,
(B) is a correspondence relationship when the lip recognition result is prioritized. Which recognition result is adopted is determined according to factors such as the surrounding environment in which this game device is operated, and it is possible to register this in the game device in advance, Alternatively, the game device may be configured such that the operator inputs it by himself. For example, as shown in (a), the voice recognition result is prioritized when a healthy person who does not hinder the utterance and the surrounding noise is relatively small. If the noise is very loud, adopt (b).

The integrated judgment section 123 outputs the integrated judgment result determined as described above as a signal 15. Finally, the control unit 6 outputs a radio control signal according to the determination result to control the airship 7.

As described above, according to this embodiment, the movement of the lips is recognized together with the voice signal, and the results of both are recognized in an integrated manner. Command) can be recognized. At the same time, it also has the effect of enabling a person with speech disabilities to use the game by voice operation. Further, as in the first and second embodiments described above, since the movement of the lips is detected by the combination of the LED 21 and the photodiode 22, it is much cheaper than the method of capturing the image of the lips using a video camera or the like. There is also an effect that can be realized.

Although a detailed description is omitted, in the present embodiment as well as in the second embodiment, the user of the game registers a standard pattern for lip recognition, but it corresponds to an unspecified speaker beforehand. It is also possible to prepare a standard pattern in a form that allows the user to omit registration by the user.

Further, in the above-mentioned first to third embodiments, the game device for controlling the airship 7 by the radio control signal is described as an example, but the game device to which the present invention can be applied is not limited to this. Is. For example,
By providing the number of operators as described in any of the above embodiments, it is possible to realize a game device in which a plurality of operators can play at the same time.

The input device of the present invention will be described below. FIG.
FIG. 5 is a diagram simply showing the configuration of the input device of the present invention.
The input device of the present invention includes a headset 154, a support 155 attached to the headset 154, and a base 153 provided with a photodiode 151 and an LED 152, and the base 153 is joined to the support 155 at a predetermined angle. (See FIG. 15A). By adjusting the angle between the base 153 and the support 155, it is possible to change the direction in which the light emitted from the LED 152 is applied to the lip portion of the operator. This input device irradiates the operator's lips with the light emitted from the LED 152 and reflects the reflected light into the photodiode 151.
It is a device for inputting the movement of the lips by detecting with. Such an input device is, for example, the above-mentioned first to third embodiments.
Can be used as an image input unit in. Also,
If a microphone 156 is added to the stand 153 ((b) of FIG. 15)
This input device can also be used as a voice input device.

An input device having no microphone as shown in FIG. 15A can be used as the image input section of the second embodiment. Further, the input device having a microphone as shown in FIG. 15B can be used as a device that serves as both the voice input unit and the image input unit in the first and third embodiments.

As described above, the input device of the present invention uses the photodiode 151, the LED 152, and the microphone 156, which are very small in size and can be mounted very lightly. The weight is very small. Further, since all the constituent elements used are inexpensive, it can be realized at low cost. Further, the input device of the present invention may be used in headset 154.
Since it is fixed to the operator's head by this, the positions of the lips and the photodiode 151 and the LED 152 can be made substantially constant. Therefore, the movement of the lips can be stably input. Further, since the input device of the present invention inputs the movement of the lips by light, converts it into an electric signal and outputs it, a conventional input device, for example, a device for inputting an image instead of the movement of the lips, or an ultrasonic wave is used. It is possible to make the configuration simpler than that of the input device, which has a large and complicated configuration such as a device to be used.

In this example, the photodiode and LE
Although only one D is mounted, a plurality of Ds can be mounted. For example, LED
By preparing two sets of photodiodes and arranging each set in a cross shape, it is possible to detect the direction of movement on the surface.

As described above, according to the present invention, it is possible to obtain a game device which can be operated by a voice which is natural to humans and which requires no operation proficiency.
Moreover, since the movement of the lips is used instead of recognizing the words (commands) input from only the voice, stable operation is possible even under noise. Furthermore, since the movement of the lips is detected by the combination of the LED and the photodiode (phototransistor), it can be realized at a low cost as compared with the case of using a video camera or ultrasonic waves.

Further, as described in the first embodiment, the utterance section of the speaker is detected from the movement of the lips, and this is used as the judgment material of the speech recognition result. Can be prevented. In addition, as described in the second and third embodiments, if the airship is controlled by recognizing the words (commands) input from the movement of the lips, the voice will be audible even in the noise. It is also possible to use it in situations where it is difficult to put out or people with speech disabilities.

Further, in the input device of the present invention, an inexpensive light emitting element (LED or the like) and an inexpensive light receiving element (photodiode or the like) are attached to a light headset, a support and a stand. Therefore, an extremely light and inexpensive input device can be realized.

In the first to third embodiments, the movement of the object is controlled according to the recognized voice or movement of the lip. However, the movement of the object controlled based on the movement of the voice or the lip is not limited to movement, but may be movement such as saying back some words. Described below are various devices for causing an object to perform some action (including movement) according to the recognized voice.

An apparatus for causing an object to perform some action in accordance with the recognized voice will be described below in each embodiment.

(Fourth Embodiment) In this embodiment, an apparatus for selecting one output voice from a set of output voices prepared for the voice and outputting the output voice according to the recognized voice is provided. explain.

FIG. 16 briefly shows the configuration of the voice selection device 100 of this embodiment. The voice selection device 100 includes a random number generation unit 101, a selection unit 102, an input / output state memory 103, a state transition unit 104, and an input / output state database 105. The input / output state database 105 stores a plurality of input / output state tables in advance. Each input / output state table includes an input x in the state s (x is a non-negative integer) and a set sp (x, i) of n (s) output voices for the input x (0 ≦ i <n (s)). Includes and. FIG. 17 shows an example of the input / output state table. The input / output state memory 103 initially stores an initial state table 201 shown in FIG. Random number generator 10
1 determines the i used to select one voice to output from the set of output voices.

The operation of the voice selection device 100 will be described below. When the selection unit 102 has an input x from the outside, the selection unit 102 refers to the input / output state table stored in the input / output state memory 103, and outputs the output voice set sp (x, i) corresponding to the input x. select. Then, the selection unit 10
2, the random number generator 101 determines the random number r (n (s)) (where 0 ≦ r (n (s)) <n (s)), and outputs the output voice as i = r (n (s)). Set sp (x,
Select one voice from i). Then, this is output to the outside.

The output from the selection unit 102 is given not only to the outside but also to the state transition unit 104. Selection unit 102
Upon receiving the output from, the state transition unit 104 refers to the input / output state database 105 and rewrites the contents of the input / output state memory 103 into the input / output state table for the output from the selection unit 102. For example, initial state 2
When 01 is output, the state transition unit 104 refers to the input / output state database 105,
The table of the input / output state 202 for the output "Is it fine?" Then, the table of the extracted state 202 is stored in the input / output state memory 103.

In this way, the voice selection device 1 of this embodiment
00 outputs a voice selected using a random number with respect to the input voice. Therefore, a simple dialogue system can be constructed by using the voice selection device 100. Further, as shown in FIG. 18, by using the voice selection device 100a having a simple configuration in which the state transition unit 104 and the input / output state database 105 are omitted, it is possible to make a one-time response to the input voice. You can also

The voice selection devices 100 and 100a.
27 can be used in combination with the voice recognition device 1201 as the voice selection device 1202 of the voice reaction device 1203 as shown in FIG. Specifically, first, when a voice is recognized by the voice recognition device 1201, the recognition result is input to the voice selection device 1202 by an identification number attached to the voice, for example. Voice selection device 12
02 uses input identification number as input x, randomly selects one voice from the output voice set, and outputs it. As a result, when a certain voice is input, a voice corresponding to the voice is output, and further, the voice reaction device 1203 that can make various responses to the same input voice can be realized. For example, the voice selection device 1202
When the voice recognition device 1201 outputs the voice "Ohayo" as the recognition result in the initial state, the identification number 1 given to the voice "Ohayo" is input as the input x to the voice selection device 1202. (Fig. 2 (a)
reference). In response to this, the voice selection device 1202 causes the set sp including the two output voices of “Good morning” and “How are you?”
One is randomly selected from (1, i) and output.

In this voice reaction device 1203, it is necessary to register a voice that can be accepted as an input in the voice selection device 1202, prior to the actual operation. When a voice not included in the registered voice set is input to the voice selection device 1202, for example, a voice “what?” May be output from the voice selection device 1202. When the device of the third embodiment is used as the voice recognition device 1201, when the reliability of the recognized voice is low, the voice for selecting the voice again is input to the voice recognition selection device 12.
It is also possible to output from 02.

As described above, in the voice selection device of the present invention, a plurality of tables indicating the input / output states are prepared and the input / output states are changed according to the past input / output history. Therefore, by using the voice selection device of the present invention, it is possible to realize a device for performing a simple dialogue. Further, this voice selection device has a plurality of output voice candidates for one input, and randomly selects and outputs one of these output voice candidates. Therefore, it is possible to obtain a voice reaction device capable of giving a varying response instead of always giving the same response to one input.

(Fifth Embodiment) Next, a direction detecting device and a direction selecting device of the present invention will be described.

First, the direction detecting device 400 will be described with reference to FIG. The direction detection device 400 includes a direction detection unit 401 and a plurality of microphones 402 connected to the direction detection unit 401, and the microphone 402 is attached to an object to be controlled. Here, the operation of the direction detection device 400 will be described by taking the case where the number of microphones is four as an example. When voice is input from the four microphones m (i) (i = 0, 1, 2, 3), the direction detection unit 401 causes
The input voice sp (m (i), t) is converted into the frame f (m
(I), j) 501 (0 ≦ j). For example, the length of one frame is 16 ms. Next, the direction detector 40
1 obtains the energy e (m (n), j) of the voice in each frame for each frame, and obtains the obtained energy e
(M (n), j) are sequentially stored in a circular memory (not shown) having a length 1 (for example, length 100). At this time, the direction detection unit 401 obtains the sum of the energy of the past 1 frame for each microphone every time the energy of each frame is stored, and determines the microphone having the maximum energy sum.
Subsequently, the direction detection unit 401 compares the maximum value of the energy sum with a threshold value The that is experimentally determined in advance, and if the maximum value of the energy sum is greater than the threshold value The, the direction detection unit 401 moves to the microphone. It is determined that the direction to reach is the direction in which the sound is heard. The microphone number i thus determined is the direction in which the voice is input, and the direction detection unit 4
Output from 01.

Direction detecting device 400 operating in this way
28 is used in combination with the operating device 1302 as shown in FIG. 28, for example, it is possible to configure a voice reaction device 1303 that performs a predetermined operation according to the direction in which the voice is heard. Specifically, for example, if an operation device 1302 and a direction detection device 1301 (400 in FIG. 19) for moving the object are attached to the object (for example, a balloon or a stuffed animal), the object moves to a person who makes a human voice. As described above, it is possible to make a device that performs a predetermined operation in the direction in which the sound is heard according to the sound.

As an example of the above-mentioned operating device 1302, the operating device 1302 has three motors with propellers attached to an object and a driving device for these motors, and the direction to move next is set. There is a device that controls the three motors so that when input, the object moves in that direction.

Next, the direction selection device will be described with reference to FIG. The direction selection device 600 includes an offset calculation unit 6
01, an azimuth meter 602, and a target direction memory 603, and can be used as a device for controlling the moving direction of an object or the direction of an object. The offset calculation unit 601 inputs x (where x is a non-negative integer) indicating the direction in which the object should move next or the direction in which the object should face.
Is input, an offset corresponding to the input x is output based on the table stored in advance in the offset calculation unit 601. The output offset is the compass 602
It is added to the actual direction of the target object measured at that time and sent to the target direction memory 603. The target direction memory 603 stores a value obtained by adding an offset to the actual direction from the azimuth meter 602, as the direction in which the next object should move or the direction in which the next object should face.

As described above, the direction selecting device of FIG. 21 is used to change the direction of an object according to the input x, with the direction in which the object is currently moving or the direction in which the object is facing as a reference. .

If the direction selecting device 700 shown in FIG. 22 is used instead of the direction selecting device 600 shown in FIG. 21, the direction of the object is not changed to a relative direction based on the current direction, but an absolute direction. You can change in any direction. In the direction selection device 700 of FIG. 22, when the direction calculation unit 701 receives an input x (x is a non-negative integer) indicating an absolute direction (for example, north) from the outside, it outputs a value corresponding to the input x. To do. The output value is stored in the target direction memory 603 as it is as the target direction. This direction calculation unit 7
Similarly to the offset calculator 601 described above, 01 can also be realized by holding the values of the absolute direction with respect to the input x as a table. After storing the target direction in the memory 603 in this way, the direction selection device 700
Measures the current direction in sequence while the object is moving or changing direction, and outputs the difference between the measured direction and the direction stored in the target direction memory 603. To do. If feedback control is performed on the object based on this output, the object can be moved in the intended absolute direction, or the direction of the object can be changed.

When the direction selection device as described above is combined with a voice recognition device and an operation device, as shown in FIG. 29, if the direction or the moving direction of the object is input by voice, the direction of the object will be changed accordingly. Alternatively, it is possible to realize the voice reaction device 1402 in which the moving direction changes. In the voice reaction device 1402, the voice recognition device 1201
Is input to the direction selection device 1401 and the output of the direction selection device 1401 is input to the operation device 1302. This makes it possible to control the operation of the target object while comparing the current direction or moving direction of the target object with the target direction.

Consider, for example, a case where the object is currently facing the direction of 0 degree when the north is 0 degree and the east direction is the positive direction. At this time, the direction selecting device 600 (see FIG. 21) is used as the direction selecting device 1401. When the voice recognition device 1201 recognizes the voice indicating the target direction as the word “right”, the offset calculation unit 601 of the direction selection device 600 associates +90 degrees with the word “right”. If the table is stored, the direction selection device 600 sends an output to the operation device 1302 instructing the operation device 1302 to change the orientation or movement direction of the object by 90 degrees eastward from the current orientation. At this time, by the direction selection device 600,
The current direction and the target direction are constantly compared during the change of the direction or the moving direction of the object. Operating device 1302
Is controlled so that the direction or movement direction of the object changes to a target direction according to the output of the direction selection device 600. Alternatively, when the direction selecting device 1401 is the direction selecting device 700 of FIG. 22,
As a word indicating the target direction, words such as "north" and "southwest" are input instead of "right" and "left". At this time, the direction selection device 700 sets 0 degree if the input word is “north”,
If it is "southwest", -135 degrees is stored in the target direction memory as the target absolute direction, and the above-described operation is performed. The target direction here is -180 degrees to +1.
80 degrees.

The direction detecting device and the direction selecting device of this embodiment may be combined with the operating device. in this case,
As shown in FIG. 30, the detection result of the direction detecting device 1301 is input to the direction selecting device 1401, and the direction selecting device 14
The output of 01 is input to the operating device 1302. Thus, the voice reaction device 1501 that changes the direction or moving direction of the target object to the direction in which the sound is heard while comparing the current direction or moving direction of the target object with the target direction. Can be realized.

(Sixth Embodiment) In this embodiment, an apparatus relating to voice recognition will be described. As shown in FIG. 26, this device includes a voice end point detection device 1101 and a voice detection device 11
02, feature quantity extraction device 1103, distance calculation device 1104
And a dictionary 11105.

First, a voice end point detection device 1101 that receives a signal corresponding to an input voice and detects a voice end point based on the signal will be described. In the present specification, the “voice end point” means the time when voice input ends.

The voice end point detection device 1101 of this embodiment
Is connected to a voice input device such as a microphone. When the voice s (t) is input from the voice input device, the voice end point detection device 1101 converts the input voice s (t) into a frame f (i) (i is a non-negative integer) as shown in FIG. To obtain energy e (i) in each frame.
In FIG. 23, the voice s (t) is represented by the curve 801, and the energy e (i) is represented by the curve 802. Next, the voice end point detection device 1101 obtains the variance of energy from the frame to a frame before a predetermined number of frames each time a frame of voice is input, and compares it with a threshold Thv that is experimentally determined in advance. . As a result of the comparison, if the energy dispersion intersects with the threshold Thv from the larger side to the smaller side, the intersection point is determined as the voice end point.

Here, a method of obtaining the variance from the energy of each frame for a certain period will be described. First, although the method of using the circular memory is used, the energy obtained for each frame is sequentially stored in the circular memory 803 of length 1. Then, every time the energy of one frame is obtained, the energy of the frame traced back for a certain period from the energy is circulated.
The variance is obtained by referring to the data from 03.

There is also a method of obtaining the energy distribution without using a circular memory. In this method, the speech end point detection device 1101 holds the average m (i-1) and the variance v (i-1) for a predetermined number of past frames, and the energy e (i ), The weighted sum of the newly obtained energy e (i) and the past average energy m (i-1) is set as the new energy average m (i), and the past variance v The weighted sum of (i-1) and | e (i) -m (i) | is set as a new variance v (i). In this way, the pseudo energy distribution can be obtained. Here, the damping constant α is used for weighting, and a new mean and variance are obtained using the following equation. 1.02 is used as α.

[0124]

[Equation 1]

By doing so, a circulating memory is not required, which leads to a memory saving, saves time and labor such as calculating the total energy within a fixed period each time new energy is obtained, and shortens the processing time. Connect

Next, the voice detection device 1102 for extracting the section in which the voice is actually pronounced will be described. Circular memory 80 for storing energy for extraction of this section
In addition to 3, the circulating memory 902 for storing the smoothing energy is prepared. As shown in FIG. 24, the energy 802 is stored in the memory 803 and the memory 902 is stored in the memory 902 every time the energy of one frame is obtained. Energy 9
01 is accumulated. Speech end point 90 as described above
When 3 is obtained, the history of energy and smoothing energy remains in these circular memories 803 and 902, and the length 1 of these circular memories is set to a sufficient length (for example, a length corresponding to 2 seconds). ), You can leave one word of energy. Therefore, the voice detection device 1102 extracts the section in which the voice is pronounced using the energy and the smoothing energy stored in these memories.

The extraction of the section is performed in the following procedure.
First, the threshold Th is determined as described later. This threshold value Th and the energy stored in the circulation memory 803 are compared in order from the past one, and the point at which the energy exceeds the threshold value for the first time is set as the start point of the section where the sound is produced. On the contrary, the point at which the energy first crosses the threshold value when going back from the voice end point to the past is set as the end point of the section where the voice is pronounced. In this way, the section in which the voice is pronounced is extracted.

Here, how to determine the threshold Th will be described.
First, the maximum energy value max1001 in the memory 803 at the time when the voice end point is detected, and the memory 902.
The smoothing energy minimum value min1002 is calculated. The threshold value Th is calculated from the following equation using these values.

[0129]

[Equation 2]

However, a value of about 0.07 was adopted as β.

Also, here, as a method of smoothing the energy, a method of taking a median value within a fixed window is used. However, the smoothing method is not limited to this, and for example, an average value may be taken.
It should be noted that the maximum value of the energy was used instead of the maximum value of the smoothing energy when the threshold value Th was calculated, because when the maximum value of the smoothing energy was used to calculate the threshold value Th, the word length varied. This is because the maximum value fluctuates significantly and the threshold Th also fluctuates accordingly, and as a result, good voice detection cannot be performed. Further, since the minimum value of the smoothing energy is used for calculating the threshold Th, it is possible to prevent noise other than voice from being detected.

As described above, the voice detecting device 1102 extracts the section in which the voice is sounded, that is, detects the portion of the input signal corresponding to the voice.

Next, the feature quantity extraction device 1103 extracts a feature quantity for recognition from the detected voice.
Similar to energy, the feature quantity is also calculated for each frame and stored in the circulation memory. Here, the feature amount is a feature amount vector including three elements of the difference between frames, which is the logarithm of the zero crossing number of the original signal, the zero crossing number of the differential signal of the original signal, and the energy of the original signal.

In this way, the voice end point detection device 1101,
Speech detection device 1102 and feature amount extraction device 1103
The voice feature amount vector obtained through the above is input to the distance calculation device 1104. The distance calculation device 1104 collates each of a plurality of voice feature amount vectors registered in advance in the dictionary 1105 with the input feature amount vector, and outputs the one having the best score as the recognition result. The matching method may simply take the Euclidean distance between the vectors, or the DP matching method may be used.

As described above, the apparatus of this embodiment performs voice recognition. This voice recognition device can be used in combination with the voice selection device 1202 described in the fourth embodiment as shown in FIG. 27, or the direction selection device 1401 and the operation described in the fifth embodiment as shown in FIG. Device 1
It can also be combined with 302. It is also possible to form a voice reaction device 1601 which moves the entire device in a target direction by using the result of the voice recognition device 1201 as an input of the operation device 1302 simply in combination with the operation device 1302.

Further, among the voice reaction devices described in Embodiments 4 to 6, the voice recognition device 1201 is included, and the signal transmission device 1701 is added to the voice recognition device side. Voice selection device 12 coming in the latter stage
02, the direction selection device 1401 and the operation device 1302, the signal receiving device 1702 can be added to remotely control the object by using only the voice recognition device as a remote controller. Here, it is possible to use infrared rays or wireless for signal transmission / reception.

By attaching the above-described voice reaction device to a balloon, it becomes possible to interact with the balloon and control the balloon, and it becomes possible to make a toy that makes the best use of the warmth unique to the balloon.

Further, as shown in FIG. 33, a voice reaction device 120 equipped with the above-mentioned voice recognition device and voice selection device.
If two objects with 3 attached to the balloon 1801 are prepared, and two voice reaction devices are configured to interact with each other instead of talking to the voice reaction device by a person, a toy that interacts freely It becomes possible to make. Furthermore, prepare a plurality of balloons 1801 with the voice reaction device,
It is also possible to have a dialogue. At this time, if each balloon with a voice reaction device has a reject function in the voice recognition process, it becomes possible to respond only to a specific word, and only one balloon responds to a certain utterance. It is also possible to configure. For example, it is possible to give a name to each balloon 1801 and make it react only when the name is called. Here, the reject method calculates a distance with an internal dictionary when performing voice recognition, but there is a method in which a threshold value is experimentally determined and one that exceeds the threshold value is rejected. Further, by incorporating a clock into the voice reaction device, and when a predetermined time has elapsed, one voice is randomly selected from the registered output voice set and outputted, so that the voice reaction device side starts the dialogue. It is also possible to construct a toy capable of playing.

The object is not limited to a balloon, and may be a stuffed animal, a doll, a photograph or a picture. It may also be a moving image on the display. Further, as the object, an anti-gravity device other than a balloon (for example, a device that is levitated by a propeller like a helicopter, or a device that is levitated by magnetic force like a linear motor car) may be used.

[0140]

As described above, according to the present invention,
It is possible to obtain a game device that can be operated by a voice that is natural to humans and does not require operation proficiency. Also, instead of recognizing words (commands) input from only voice, it uses the movement of the lips,
Stable operation is possible even under noise. Furthermore, since the movement of the lips is detected by the combination of the LED and the photodiode (phototransistor), it can be realized at a low cost as compared with the case of using a video camera or ultrasonic waves.

Further, in the voice recognition device of the present invention, the utterance section of the speaker is detected from the movement of the lips, and this is used as the judgment material of the voice recognition result, so that the erroneous recognition due to the utterance of the person other than the speaker is prevented. You can In another voice recognition device of the present invention, a word (command) input from the movement of the lips
In order to control the airship by recognizing the above, it is possible to use even in a noisy situation, in a situation where it is difficult to make a voice, or for a person who has a speech disorder.

In the input device of the present invention, an inexpensive light emitting element (LED or the like) and an inexpensive light receiving element (photodiode or the like) are attached to a light headset, a support and a stand. Therefore, an extremely light and inexpensive input device can be realized.

As described above, the voice selection device of the present invention prepares a plurality of input / output states and changes the input / output states according to the past input / output history. Therefore, it becomes possible to provide a device for a simple dialogue by using this voice selection device. Further, the voice selection device of the present invention prepares a plurality of outputs for one input and outputs one randomly selected from these, so that the same response is not always given to one input, but a change of You can give a response.

Further, the direction detecting device of the present invention inputs voice by a plurality of microphones and detects the microphone having the maximum energy. Thereby, the direction in which the voice is uttered can be detected. Further, by using the direction selection device of the present invention, while detecting the current position by the azimuth meter, the object can be moved accurately in the input direction, or the direction of the object can be changed to the input direction. can do.

Further, in the voice recognition apparatus of the present invention, the voice end point detection apparatus first obtains a rough voice end point, and then the voice detection apparatus automatically obtains a threshold value. Here, since the threshold value is determined from the maximum value of the energy of the input voice and the minimum value of the smoothed energy,
Good voice segment extraction can be performed regardless of the length of the vocal segment of the voice. When the voice detection device detects a voice using a threshold, a feature amount is obtained from this voice, and voice recognition is performed based on the feature amount.

Various voice reaction devices can be obtained by appropriately combining the above-mentioned devices. For example, if you combine a voice recognition device and a voice selection device,
I get a voice reaction device that responds when a person speaks in a voice,
This makes it possible to build a man-machine interface. Also, if the direction detection device and the operation device are combined, it becomes possible to move the object in response to the voice, and if the voice recognition device, the direction selection device and the operation device are combined, the content of the voice is shown. It is possible to accurately move the target object in the direction and to change the direction of the target object in the direction indicated by the voice content. Furthermore, if a signal transmission device is connected to the voice recognition device of the voice reaction devices, and a signal reception device is connected to a device that comes after the voice recognition device and attached to an object, a voice that can be remotely operated is possible. A reactor can be realized.

Furthermore, by preparing a plurality of voice reaction devices as described above, it is possible to construct a toy that automatically interacts with each other. Also, if you attach each voice reaction device to the balloon, you will have the warmth peculiar to the balloon,
Moreover, it is possible to make toys that can talk to you. It is also possible to build a voice reaction device in which a human being speaks instead of speaking from a human by incorporating a clock and outputting an appropriate voice at a certain time.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a game device according to a first embodiment of the present invention.

FIG. 2 is a diagram showing a detailed configuration of an image input unit according to first to third embodiments of the present invention.

FIG. 3 is a diagram showing a detailed configuration of a vocalization section detection unit in the first exemplary embodiment of the present invention.

FIG. 4 is a block diagram showing a detailed configuration of an integrated determination unit in the first exemplary embodiment of the present invention.

FIG. 5 is a graph showing an output example of differential signals in the first to third embodiments of the present invention.

FIG. 6 is a diagram for explaining the processing operation of the utterance section detection unit of FIG.

FIG. 7 is a diagram for explaining the processing operation of the integrated determination unit of FIG.

FIG. 8 is a block diagram showing a configuration of a game device according to a second embodiment of the present invention.

FIG. 9 is a block diagram showing a detailed configuration of a lip recognition unit in second and third embodiments of the present invention.

FIG. 10 is a diagram showing the processing operation of the differentiating circuit in the second and third embodiments of the present invention.

FIG. 11 is a diagram showing the processing operation of the pattern matching unit according to the second and third embodiments of the present invention.

FIG. 12 is a block diagram showing a configuration of a game device according to a third embodiment of the present invention.

FIG. 13 is a diagram showing a processing operation of an integrated determination unit in the third embodiment of the present invention.

FIG. 14 is a diagram showing a processing operation of an integrated determination unit in the third exemplary embodiment of the present invention.

FIG. 15 is a diagram showing a specific configuration example of the input device of the present invention.

FIG. 16 is a diagram showing a configuration of a voice selection device according to a fourth exemplary embodiment of the present invention.

17 is a diagram showing an input / output state in the voice selection device of FIG.

FIG. 18 is a diagram showing a configuration of a voice selection device of a modified example of the invention.

FIG. 19 is a diagram showing a configuration of a direction detection device according to a fifth exemplary embodiment of the present invention.

FIG. 20 is a diagram illustrating a waveform and a frame of input voice.

FIG. 21 is a diagram showing a configuration of a direction selection device according to a fifth exemplary embodiment of the present invention.

FIG. 22 is a diagram showing the configuration of another direction selection device of the fifth exemplary embodiment of the present invention.

FIG. 23 is a diagram illustrating a speech waveform, energy, and a circular memory.

FIG. 24 is a diagram illustrating a method of detecting a voice end point according to the sixth embodiment of the present invention.

FIG. 25 is a diagram illustrating a voice detection method according to a sixth embodiment of the present invention.

FIG. 26 is a block diagram showing a configuration of a voice recognition device according to a sixth embodiment of the present invention.

FIG. 27 is a diagram showing a configuration of a voice reaction device using a voice recognition device and a voice selection device of the present invention.

FIG. 28 is a diagram showing a configuration of a voice reaction device using the direction detection device and the operation device of the present invention.

FIG. 29 is a diagram showing a configuration of a voice reaction device using the voice recognition device, the direction selection device, and the operation device of the present invention.

FIG. 30 is a diagram showing a configuration of a voice reaction device using a direction detection device, a direction selection device, and an operation device of the present invention.

FIG. 31 is a diagram showing a configuration of a voice reaction device using the voice recognition device and the operation device of the present invention.

FIG. 32 is a diagram showing the configuration of a remotely operable voice reaction device of the present invention.

FIG. 33 is a diagram showing an example of a toy using the voice reaction device of the present invention.

FIG. 34 is a diagram showing a configuration of a conventional game device.

[Explanation of symbols]

 1 voice input unit 3 image input unit 2 voice recognition unit 4 vocalization section detection unit 5, 123 integrated judgment unit 6 control unit 7 airship 21 LED 22 photodiode 81 lip recognition unit 100, 100a voice selection device 101 random number generation unit 102 voice selection Part 103 Input / output state memory 104 State transition part 105 Input / output state database 400, 1301 Direction detection device 401 Direction detection part 600, 700, 1401 Direction selection device 601 Offset calculation device 602 Direction meter 603 Target direction memory 701 Direction calculation device 1101 Voice End point detection device 1102 Speech detection device 1103 Feature amount extraction device 1104 Distance calculation device 1105 Dictionary 1201 Speech recognition device 1202 Speech selection device 1302 Operating device 1701 Signal transmission device 1702 Signal reception device

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification number Reference number within the agency FI Technical display area G10L 9/12 301 B H04Q 9/00 301 B (72) Inventor Kazuhiro Kayashima Daimon Kadoma, Kadoma City, Osaka Prefecture 1006 address Matsushita Electric Industrial Co., Ltd.

Claims (39)

[Claims] 1. Inputting at least one voice including a voice generated by an operator, and first inputting the input voice.
And a voice recognition means for recognizing the at least one voice based on the first electric signal output from the voice input means. An image input means for optically detecting the movement of the operator's lip, converting the detected movement of the lip into a second electric signal, and outputting the second electric signal; A generated section detecting unit that receives an electric signal and obtains a section in which the voice is generated by the speaker based on the received second electric signal, and the at least one voice recognized by the voice recognizing unit. And an integrated determination unit that extracts the voice generated by the operator from the at least one voice based on the period obtained by the generation period detection unit, and the integrated determination unit extracted by the integrated determination unit. voice Based on,
A game device comprising: a control unit that controls an object. 2. The voicing section detecting means includes a differentiating means for detecting a degree of change of the second electric signal output from the image inputting means, and a predetermined degree of the change detected by the differentiating means. The game device according to claim 1, further comprising: a unit that determines that the corresponding voice is generated by the operator when the value exceeds. 3. The integration judgment means adds an interval of a predetermined length to the interval obtained by the vocalization interval detection means to create an evaluation interval, and the speech recognition means recognizes the evaluation interval. At least 1
A unit that detects the recognition result output time of one voice output from the voice recognition unit, compares the recognition result output time with the evaluation section, and outputs the recognition result output time of the at least one voice. 3. The game apparatus according to claim 1, further comprising: a unit that determines that a voice included in the evaluation section is the voice uttered by the operator. 4. The movement of the operator's lips is optically input,
An image input unit that converts the input movement of the lip into an electric signal and outputs the electric signal, and obtains the movement of the lip based on the electric signal, and displays a word corresponding to the obtained movement of the lip. A game device comprising: a lip recognition unit that recognizes and outputs a recognition result; and a control unit that controls an object according to a control signal based on the recognition result. 5. The lip recognizing means selects a storage means for storing a predetermined number of words, selects one from the predetermined number of words according to the obtained movement of the lip, and selects the selected word. The game device according to claim 4, further comprising a matching unit that determines that the word is the word corresponding to the movement of the lip. 6. The storage means stores movements of lips corresponding to the predetermined number of words as a standard pattern, and the matching means stores the movements of the lips obtained for all of the standard patterns. 6. The game device according to claim 5, wherein the word corresponding to one of the standard patterns having the smallest distance is selected. 7. A voice input means for inputting voice, converting the voice into another electric signal, and outputting the other electric signal, and based on the other electric signal output from the voice input means. An integrated judgment means for outputting the control signal to be given to the control means based on both the voice recognition means for recognizing the voice, the recognition result by the voice recognition means, and the recognition result by the lip recognition means. The game apparatus according to claim 4, further comprising: 8. A means for obtaining a voice recognition reliability with respect to the recognition result by the voice recognition means, and a means for obtaining a lip recognition reliability with respect to the recognition result by the lip recognition means. The integrated determination means selects one of the recognition result by the voice recognition means and the recognition result by the lip recognition means based on the voice recognition reliability and the lip recognition reliability, and The game device according to claim 7, wherein is output as the control signal. 9. The image input means, a light emitting means for emitting light, and a light receiving means for receiving the light reflected by the lip of the operator and converting the received light into the second electric signal. The game device according to claim 1, comprising: 10. The image input means includes a light emitting means for emitting light and a light receiving means for receiving the light reflected by the lips of the operator and converting the received light into the electric signal. The game apparatus according to claim 4, wherein 11. The image input means includes a light emitting means for emitting light and a light receiving means for receiving the light reflected by the lips of the operator and converting the received light into the electric signal. The game device according to claim 7, which is operating. 12. The game device according to claim 9, wherein the light is laterally applied to the lip. 13. The game device according to claim 9, wherein the light illuminates the lip from the front. 14. The game device according to claim 1, wherein the voice input unit has at least one microphone. 15. The voice input unit has at least one microphone, and the at least one microphone and the light emitting unit and the light receiving unit of the image input unit are provided on one stand. The game device according to claim 11. 16. A headphone-like headset, a support having one end joined to the headset, and a base joined to the other end of the support, on which
A pedestal provided with at least one light emitting element that generates light that is applied to the lips of the operator, and at least one light receiving element that receives the light reflected by the lips;
An input device equipped with. 17. The input device according to claim 16, wherein voice input means for inputting voice is provided on the table. 18. A first storage means for storing a plurality of tables, wherein each of the plurality of tables includes a plurality of words that can be output with respect to one input.
Storage means, second storage means for storing one of the plurality of tables, and one of the tables stored in the second storage means in response to an external input. Selecting one word from the plurality of words, and outputting the selected one word as a voice, and the one table stored in the second storage means, And a transition means for updating from the plurality of tables stored in the storage means to another table determined according to the selected one word. 19. The voice selection device according to claim 18, further comprising means for generating a random number, wherein the selection means selects the one word from the plurality of words by using the random number. 20. Storage means for storing a table, wherein the table includes storage means containing a plurality of words that can be output in response to one input, and input from the outside, A voice provided with a selecting means for selecting one word from the plurality of words contained in the stored table by using a random number and outputting it as a voice, and a means for generating the random number. Selection device. 21. A voice selection device according to any one of claims 18, 19 and 20, and a voice recognition means for inputting voice, recognizing the voice, and giving a recognition result to the voice selection device. And a voice reaction device. 22. A game device comprising the voice reaction device according to claim 21. 23. A game device comprising a plurality of voice reaction devices according to claim 21, whereby the voice reaction devices interact with each other. 24. A plurality of voice input units for converting an input voice into an electric signal, wherein the plurality of voice input units correspond to different directions, and the energy of the electric signal is stored in the plurality of voice input units. For each of the plurality of voice input units, one of the plurality of voice input units having the largest energy is determined, and the voice is generated in the direction corresponding to the determined one voice input unit. A game device that includes a direction detecting unit that determines that the direction is a reversed direction. 25. The apparatus further comprises an operating means for moving the object, and a control means for controlling the operating means so as to change the operating direction of the object in the determined direction. The game device according to 1. 26. The game device, wherein a measuring means for measuring a current direction of movement of an object, and the determined direction are input, and a target direction is determined based on the current direction and the determined direction. The method further includes: direction selecting means having means for storing the desired direction, and means for operating the object, the direction selecting means including the direction of the target and the current direction. 25. The game device according to claim 24, wherein the movement means is controlled so that the current direction of the movement of the object and the target direction substantially coincide with each other by using the difference in the direction. 27. Input means for inputting a relative direction by voice, measuring means for measuring a current direction of an object, and a target direction based on the current direction and the input relative direction. A game device having a direction selection means having a means for determining and storing the target direction, the direction selection means using the difference between the target direction and the current direction, A game device for controlling the target object so that the current direction and the target direction substantially match. 28. The input unit according to claim 27, further comprising an input unit for inputting the voice, and a recognition unit for recognizing the relative direction based on the input voice. Game device. 29. Input means for inputting an absolute direction by voice, means for determining a target direction based on the absolute direction and storing the target direction, and measuring the current direction of the object. A game device comprising a direction selecting means having a measuring means, wherein the direction selecting means uses the difference between the target direction and the current direction to determine the current direction and the target direction of the object. A game device that controls the object so that and substantially match. 30. The input unit has an input unit for inputting the voice, and a recognition unit for recognizing the absolute direction based on the input voice. Game device. 31. Receiving an electrical signal corresponding to voice,
First detection means for detecting, from the electric signal, a voice end point which is a time when the input of the voice is finished, and the voice in a section where the voice is inputted is uttered based on the electric signal. Second detection means for determining a vocalization section which is a section, feature quantity extraction means for creating a feature quantity vector on the basis of the vocalization section of the electric signal, and a plurality of candidate speeches created in advance. Storage means for storing a feature quantity vector; and comparing the feature quantity vector from the feature quantity extraction means with each of the feature quantity vectors of the plurality of candidate speeches stored in the storage means, A voice recognition device comprising: means for recognizing input voice. 32. The first detecting means divides the electric signal into a plurality of frames each having a predetermined length, and the energy of the electric signal for each of the plurality of frames. 32. The voice recognition device according to claim 31, further comprising: a calculating unit that obtains the result and a determining unit that determines the voice end point based on the dispersion of the energy. 33. The deciding means decides the voice end point by comparing a predetermined threshold value with the variance of the energy, and the voice end point is the threshold value when the variance of the energy is the threshold value. 33. The speech recognition apparatus according to claim 32, wherein the variance is the time when the variance matches the threshold when changing from a larger value to a smaller value. 34. The speech recognition apparatus according to claim 32 or 33, wherein the determining means uses a variance of the energies of the plurality of frames for a predetermined number of frames. 35. The second detecting means includes means for smoothing the energy of the electric signal, and first circulating storage means for sequentially storing the energy of the electric signal for each frame without smoothing the energy. Second circular storage means for sequentially storing the smoothed energy for each frame, and the smoothing stored in the first cyclic storage means when the voice end point is detected. Threshold value calculation means for calculating a threshold value for vocalization interval detection using both the unenhanced energy and the smoothed energy stored in the second circulation type storage means, and the unsmoothed energy 33. The speech recognition apparatus according to claim 32, further comprising: a vocalization section determining unit that determines the vocalization section by comparing the vocalization section detection threshold value. 36. The threshold value calculation means, when the voice end point is detected, the maximum value of the unsmoothed energy stored in the first circulation type storage means, and the voice end point. 36. The voice recognition according to claim 35, wherein the threshold for voiced section detection is calculated using the minimum value of the smoothing energy stored in the second circulation type storage means at the time when is not detected. apparatus. 37. The feature amount detecting means, from the portion of the vocalization section of the electric signal, the number of zero crossings for each frame of the electric signal and zero for each frame of the signal obtained by differentiating the electric signal. The voice recognition device according to claim 35 or 36, wherein the number of intersections and the energy of the electric signal are calculated, and these are used as elements of the feature amount vector. 38. At least one voice recognition device according to claim 32, and at least one control means for controlling an object based on a recognition result of the at least one voice recognition device. , A voice reaction device comprising. 39. The voice reaction device is connected to the at least one voice recognition device,
Transmitting means for transmitting the recognition result by the at least one voice recognition device, and receiving means connected to the at least one control device for receiving the transmitted recognition result and giving it to the at least one control device 39. The voice of claim 38, further comprising: and the at least one controller and the receiving means are attached to the object, thereby enabling remote operation of the object. Reactor.