JP6573474B2 - Speech recognition system - Google Patents

Description

  The present invention relates to a speech recognition system having a speech recognition function.

  In recent years, some home electronic devices (hereinafter referred to as home appliances) are equipped with various additional functions in addition to the original functions so that the user can use them more conveniently and comfortably. As such an additional function, for example, there is a voice recognition function, and there are home appliances capable of voice operation and voice communication. With such a function, the user can use the home appliance conveniently, comfortably and enjoyably.

  Incidentally, in a device having a voice recognition function, it is important to increase the recognition rate of voice recognition. Therefore, for example, Patent Document 1 discloses a vehicle voice recognition system that increases the recognition speed by limiting the recognition words for voice recognition during operation of the device and increases the voice recognition rate by lowering the volume of the audio output. Is disclosed.

JP 2007-147962 A (published on June 14, 2007)

  However, depending on the device, the operation sound of the device may become a noise source for voice recognition, and the recognition rate for voice recognition may not increase. In addition, depending on the device, there is a case where an original operation (operation different from voice recognition) to be performed by the device is stopped, and an adverse effect on an operation recipient (object, object) may occur.

  Accordingly, the present invention has been made in view of the above-described problems, and an object of the present invention is to provide a speech recognition system capable of reducing adverse effects caused by suppressing or stopping sound generation while increasing the speech recognition rate.

  In order to solve the above-described problem, a speech recognition system according to an aspect of the present invention includes a first unit that performs reception of input speech and speech recognition of the received input speech, and a first unit that performs an operation that generates sound. In the voice recognition system comprising two units, when the start instruction to accept the input voice is detected, the operation is performed based on at least one of a sensing result of the sensor of the second unit or a control state of the second unit. A determination unit that determines whether or not there is urgency, and an operation control unit that suppresses or stops the operation for a predetermined period including a reception period of the input voice when the determination unit determines that there is no urgency. It is characterized by that.

  According to one aspect of the present invention, it is possible to perform processing in consideration of adverse effects caused by suppressing or stopping the operation by determining whether or not the operation of generating the sound of the second unit is urgent. That is, it is possible to prioritize the reception of the input voice when the operation that generates the sound is not urgent and to perform the process that prioritizes the operation when the operation that generates the sound is urgent. Furthermore, when it is determined that there is no urgency of the operation that generates the sound, the sound recognition rate can be increased by suppressing or stopping the operation that generates the sound for a predetermined period including the reception period of the input sound. In addition, when the operation that generates the sound is not urgent, the operation that generates the sound is suppressed or stopped, so that it is possible to reduce adverse effects caused by the suppression or stop of the operation.

  As can be seen from the above, according to one embodiment of the present invention, it is possible to provide a speech recognition system that can increase the speech recognition rate and reduce adverse effects caused by suppression or stop of the operation that generates sound.

It is a schematic block diagram of the speech recognition system which concerns on Embodiment 1 of this invention. It is a schematic block diagram of the speech recognition system which concerns on Embodiment 2 of this invention. It is a figure which shows the usage example of the speech recognition system which concerns on Embodiment 2 of this invention. It is a schematic block diagram of the speech recognition system which concerns on Embodiment 3 of this invention.

Embodiment 1
Hereinafter, an embodiment of the present invention will be described in detail with reference to FIG. In this embodiment, a case where the speech recognition system according to the present invention is applied to a home electronic device (home appliance) will be described.

(Composition of home appliances)
FIG. 1 is a block diagram illustrating a configuration of a home appliance (voice recognition system, first unit, second unit) 10 according to the present embodiment. The type of home appliance 10 is not limited as long as it is a device that performs an operation of generating sound (operation sound) while receiving input sound and performing speech recognition on the received input sound. In the following, the process of accepting input speech and recognizing speech is referred to as speech processing. An operation sound is a sound generated with an operation.

  The home appliance 10 may be, for example, a refrigerator, an air conditioner (air conditioner), an air purifier, a washing machine, a heating cooker, a cleaning robot, or the like. Further, one home appliance 10 may execute a plurality of operations that generate an operation sound. For example, if the home appliance 10 is a refrigerator, a refrigeration operation (compressor operation) or an operation of dropping ice by an ice maker into an ice tank corresponds to an operation sound. These are examples, and operations other than these may be used as operations that generate operation noise in the refrigerator.

  As shown in FIG. 1, the home appliance 10 includes a voice input unit 11, a voice output unit 12, a control unit 13, a data storage unit 14, an operation unit 15, a sensor unit 16, and a display unit 17.

  The voice input unit 11 is a voice input device such as a microphone, and the voice output unit 12 is a voice output device such as a speaker.

  The control unit 13 is a block that controls the operation of each unit of the home appliance 10. The control unit 13 includes a computer device including an arithmetic processing unit such as a CPU (Central Processing Unit) and a dedicated processor, for example. The control unit 13 comprehensively controls the operation of each unit of the home appliance 10 by reading and executing a program for performing various controls in the home appliance 10 stored in the data storage unit 14. The control unit 13 controls each block of the home appliance 10 so that the control state of the home appliance 10 corresponds to the operation mode. As the operation mode, for example, there are a dialogue mode in which a voice dialogue is performed with the user, various operation modes (weak operation mode, strong operation mode) and the like. Various operation modes are provided depending on the home appliance 10. For example, a reservation acceptance mode in which the start of operation is waited until the reservation time may be used.

  Each operation mode is started by pressing down a button (not shown) provided on the home appliance 10. Or you may start by the input of a specific audio | voice. Here, the home appliance 10 stores in the data storage unit 14 information indicating which operation mode the device is in, that is, mode information indicating the control state of the device.

  In addition, the control unit 13 has functions as an audio processing unit 131, an operation control unit 132, a determination unit 133, and a display control unit 134. These will be described later.

  The data storage unit 14 includes a RAM (Random Access Memory), a ROM (Read Only Memory), an HDD (Hard Disk Drive), and the like, and is a block that stores various information (data) used in the home appliance 10. The data storage unit 14 stores utterance data associated with the result of speech recognition by the speech processing unit 131 described below.

  The operation unit 15 is a block that executes various operations in the home appliance 10. The operation unit 15 performs an operation in which an operation sound is generated during operation. As an example of the operation unit 15, when the home appliance 10 is a refrigerator, when it is an air conditioner such as a compressor or ice machine, when the home appliance 10 is an air cleaner such as a motor that moves a compressor, a fan, or a louver, the fan or louver is moved. Motors, ion generators, etc. In addition, when the home appliance is a washing machine, a motor that rotates a drum or the like, and when it is a heating cooker, a motor that moves a fan, a magnetron, or the like. These are examples and are not limited to these.

  Here, the operation unit 15 may execute an operation that does not generate an operation sound in addition to an operation that generates an operation sound. However, in this embodiment, the operation of the operation unit 15 is not particularly specified unless otherwise specified. Suppose that the operation sound is generated.

  The sensor unit 16 is a variety of sensors included in the home appliance 10 and is used to detect an environment inside the home appliance 10 or an environment around the home appliance. The sensor unit 16 includes, for example, a temperature sensor, a humidity sensor, an illuminance sensor, a human sensor, or an imaging device (camera). However, it is not limited to these. The sensor unit 16 may include a plurality of sensors. The type of sensor included in the sensor unit 16 varies depending on the type of home appliance 10.

  The display unit 17 is a block that displays various types of information in the home appliance 10. The display unit 17 is configured as a touch panel, and may also be used as an operation unit that operates the home appliance 10.

  The voice processing unit 131 is a block that performs voice recognition on the input voice received by the voice input unit 11 and performs various processes associated with the voice recognition result. As various processes, the speech processing unit 131 reads the speech data associated with the speech recognition result from the data storage unit 14, performs speech synthesis as speech speech, and outputs speech speech from the speech output unit 12. Note that when speech data is generated as speech data, speech synthesis is not performed. Known techniques can be used for speech recognition and speech synthesis. In this case, the user can perform a voice conversation with the home appliance 10.

  In addition, as the above-described various processes, the voice processing unit 131 enters the operation mode in which the home appliance 10 corresponds to the voice recognition result when the operation mode associated with the voice recognition result is stored in the data storage unit 14. Thus, the control by the control unit 13 is executed. To give a specific example, when the home appliance 10 is a refrigerator with an ice making machine, for example, when the input voice “make ice” is voice-recognized, the operation mode associated with the result of this voice recognition is: This corresponds to the ice making mode in which the ice making machine is operated. In this case, the user can perform voice operations on the home appliance 10. The home appliance 10 may be configured to perform both voice conversation and voice operation, or may be configured to perform only one of them. Or the structure by which another process matched with the audio | voice recognition result may be performed.

  The determination unit 133 is a block that determines the presence or absence of urgency of the operation of the operation unit 15 based on at least one of the sensing result by the sensor unit 16 or the mode information of the home appliance 10. When the determination unit 133 detects a start instruction (start information) for accepting an input voice as will be described later, the determination unit 133 determines whether or not the operation unit 15 generates an operation sound with urgency. Details of this determination will be described later with a specific example. In the case where there are a plurality of operations that cause the operation sound by the operation unit 15, the determination unit 133 performs the above determination for each operation that generates the operation sound.

  The operation control unit 132 is a block that controls the operation unit 15. When it is determined by the determination unit 133 that the operation sound generated by the operation unit 15 is not urgent, the operation control unit 132 operates for a predetermined period including a reception period for receiving an input voice to the voice input unit 11. The operation | movement which the operation sound which 15 performs performs is suppressed or stopped. And after a predetermined period, suppression or a stop is cancelled | released. In addition, when the determination unit 133 determines that there is urgency, the operation control unit 132 performs the operation of the operation unit 15 for the predetermined period. Details of these controls will be described later with specific examples.

  Here, the predetermined period is a period from when the home appliance 10 detects an input voice reception start instruction (start trigger) until an input voice reception end instruction (end trigger) is detected. In the present embodiment, the start instruction is a push down of a dialogue button (not shown) provided on the home appliance 10. The dialogue button is configured such that the home appliance 10 starts accepting input voice when pressed by the user. The start instruction is not limited to the above, and may be, for example, input of a specific voice (for example, a user voice “Let's talk!” Or a sound of hitting a hand once).

  In this embodiment, the input voice reception end instruction is detected by the end of the dialog based on the content of the dialog with the user. The end instruction may be detected, for example, when there is no input sound for a certain period after outputting the speech sound from the home appliance 10. Further, the end instruction may be a push down of a dialog end button (for example, a forced end button) provided in the first unit. The dialogue end button is a button configured such that the home appliance 10 finishes accepting the input voice when pressed by the user. Alternatively, the end instruction may be input of a specific sound (for example, a user sound of “seeking” or a sound of clapping a hand twice). In addition, the control part 13 of the household appliance 10 controls from detection of the start instruction | indication of reception of input sound to detection of an end instruction | indication as a dialog mode.

  Further, the determination unit 133 determines the urgency a plurality of times during the predetermined period, and when it is determined that there is urgency, the operation control unit 132 executes the operation of the operation unit 15.

  The sensing result of the sensor unit 16 is the temperature inside the chamber if the home appliance 10 is a refrigerator, the room temperature if the home appliance 10 is an air conditioner, the sensing result of a surrounding odor sensor or gas sensor if the home appliance 10 is an air cleaner, etc. It is. If it is a result of sensing by sensor part 16, it will not be limited to these. Moreover, the result that the audio | voice input part 11 picked up the noise may be sufficient. In this case, the voice input unit 11 functions as the sensor unit 16.

  The display control unit 134 is a block that controls the display unit 17 to display various types of information. When the operation that generates the operation sound by the operation unit 15 is suppressed or stopped, the display control unit 134 causes the display unit 17 to display information indicating that.

  As described above, in the home appliance 10, the presence or absence of the urgency of the operation of the operation unit 15 is determined based on at least one of the sensing result of the sensor unit 16 or the mode information of the home appliance 10, thereby adversely affecting the suppression or stop of this operation. Can be processed. In other words, when the operation of the operation unit 15 is not urgent, it is possible to give priority to the reception of the input voice, and when the operation of the operation unit 15 is urgent, it is possible to perform processing that prioritizes this operation. .

  Further, when it is determined that the operation of the operation unit 15 is not urgent, the operation of the operation unit 15 is suppressed or stopped for a predetermined period including the reception period of the input voice. In this case, the influence of the operation sound of the operation unit 15 on the sound processing can be suppressed. That is, since the operation sound caused by the operation of the operation unit 15 that enters the input speech as noise can be suppressed, the speech recognition rate can be increased. Further, if it is determined that there is urgency during a predetermined period including the reception period of the input voice, the operation of the operation unit 15 is executed, so that no adverse effects due to the suppression or stop of the operation occur.

  As can be seen from the above, the home appliance 10 can reduce the harmful effects caused by suppressing or stopping the operation that generates sound while increasing the voice recognition rate. Therefore, it is possible to provide the home appliance 10 that is easy to use for the user and that can accurately perform voice recognition.

  As described above, when the determination unit 133 determines that there is an urgency, the operation control unit 135 executes the operation of the operation unit 15, and if the home appliance 10 is an air conditioner at that time, “the room is hot. If a message explaining the reason to the user is output or displayed, such as “I will do my best to drive from now on”, the user can easily understand.

  In addition, when the determination unit 133 determines the urgency of the operation of the operation unit 15 a plurality of times during the predetermined period and determines that there is urgency, the operation control unit 135 executes the operation of the operation unit 15. At that time, after the information for obtaining permission is output or displayed as voice information such as “Can the story be stopped by the user?”, The interactive mode may be terminated and the operation of the operation unit 15 may be executed.

  Furthermore, the determination unit 133 calculates a value indicating urgency based on at least one of the sensing result of the sensor unit 16 or the mode information of the home appliance 10, and determines that there is no urgency when the value is equal to or less than a predetermined value. If the value exceeds a predetermined value, it may be determined that there is urgency.

  Here, for example, the higher the value indicating the calculated urgency, the higher the operation control unit 132 performs the normal operation (when there is no voice reception and no voice recognition in the first unit), and the lower the operation, You may perform control which suppresses more. As a value indicating urgency, for example, when the home appliance 10 is an air conditioner, and a value indicating urgency when the set temperature has a room temperature and an opening of 5 ° C. or more, a value indicating urgency is 3 or 3 ° C. or more and less than 5 ° C. If the value indicating the urgency is set to 2, and the value indicating the urgency when there is an opening of 2 ° C. or more and less than 5 ° C. is set to 1, the following more detailed control can be performed.

  Based on the sensing result of the temperature sensor of the sensor unit 16, the cooling operation is stopped when the urgency 1 is calculated, the cooling operation is weakened when the urgency 2 is calculated, and the urgency 3 is calculated when the urgency 3 is calculated. Control that does not suppress the cooling operation can be performed. In this way, by calculating the value indicating urgency, control according to the value indicating urgency can be performed. Of course, this is just an example. In addition, the value indicating urgency may be expressed by a numerical value as described above, or may be expressed by an index that shows the degree, such as large, medium, and small. In the above description, the degree is described as three stages, but may be set more finely (more stages).

  Further, when the home appliance 10 detects an input voice reception start instruction, the voice input unit 11 functions as the sensor unit 16 and the voice input unit 11 picks up noise as a result of sensing. 15 operations may be suppressed or stopped.

  Moreover, the structure which connects the household appliance 10 to a communication network may be sufficient. In this case, you may make the management server apparatus of the household appliance 10 connected via a communication network perform a part of function of the audio | voice processing part 131. FIG. For example, the speech processing unit 131 is configured to transmit the input speech received by the speech input unit 11 to a management server device that performs speech recognition and receive a speech recognition result from the management server device as a response. Also good. When speech recognition is performed by the management server device, the cost of the home appliance 10 can be suppressed and the speech recognition function can be improved without increasing the processing capacity of the home appliance 10 and expanding the storage capacity. For example, when the management server device is connected to another information providing server device and acquires various types of information from the information providing server device, a lot of information is obtained via the home appliance 10 as a voice recognition result by the management server device. It can also be provided to the user.

  The home appliance 10 is a so-called network home appliance that is connected to a communication network and can be remotely controlled from a communication terminal device such as a tablet terminal or a smartphone, or the operation state of the home appliance 10 can be confirmed by the communication terminal device. It may be configured to function. In this case, a notification that the operation of the operation unit 15 causing the operation sound is suppressed or stopped may be transmitted to the communication terminal device as the operation state of the home appliance 10. The user of the communication terminal device that has received the notification can grasp the status of the home appliance 10 even when remote. In addition, the communication terminal device should just be comprised so that a dedicated application can be downloaded and it can communicate with the household appliance 10, for example.

  In the above description, the case where the voice recognition system according to the present invention is applied to home appliances has been described. However, the present invention is not limited to this, and the voice recognition system according to the present invention can be applied to any electronic device. Also good. For example, you may apply to a personal computer, a communication robot, etc.

(Specific example of determination by the determination unit and control by the operation control unit)
Next, control of the home appliance 10 will be described using a specific example.

(refrigerator)
When the home appliance 10 is a refrigerator, the following processing is performed. When an instruction to start accepting input voice is detected during the refrigeration operation, the operation state is set to the interactive mode, the refrigeration operation is weakened, and the rotation speed of the compressor is reduced. Return the compressor speed to its original value when the conversation mode ends. In addition, when the ice making of the refrigerator ice making machine is finished during the interactive mode, the process waits until the ice tank is dropped until the interactive mode is finished, and then drops the ice after the interactive mode is finished. Also, if the refrigerator's internal temperature rises during the interactive mode and you want to increase the compressor speed, wait for the compressor to increase until the interactive mode ends. Increase the number of revolutions. In addition, even during the interactive mode, if it is determined that the effect on the foodstuff in the refrigerator is large unless the rotation of the compressor is increased, a voice message or a display for prompting the end of the interactive mode is displayed. When the user wants to end the dialogue mode, the refrigerator detects this when a specific voice (end phrase) is uttered (speech input) or a dialogue end button is pressed to generate an end trigger. And exit interactive mode.

(Air conditioner)
When the home appliance 10 is an air conditioner, the following processing is performed. When an instruction to start receiving input voice is detected during the operation mode, the conversation mode is entered and the weak operation mode is set to reduce the rotational speed of the fan motor. Restore the number of revolutions at the end of interactive mode. Also, if the difference between the set temperature of the air conditioner and the room temperature increases during the interactive mode, it waits for the fan motor or compressor to increase until the interactive mode ends, and the interactive mode ends. Later, increase the rotation speed of the fan motor and compressor. Further, even in the interactive mode, if it is determined that the risk of heat stroke or the like is high unless the driving level is raised, a voice message or a display prompting the end of the interactive mode is issued. When the user wants to end the interactive mode, when the end trigger is generated by speaking the end phrase or pressing the dialog end button, the air conditioner detects this and ends the interactive mode.

(Air cleaner)
When the home appliance 10 is an air cleaner, the following processing is performed. When an instruction to start receiving input voice is detected during the operation mode, the conversation mode is entered and the weak operation mode is set to reduce the rotational speed of the fan motor. Restore the number of revolutions at the end of interactive mode. Also, when the air cleaner senses indoor dust or odor during the interactive mode, it waits for the fan motor speed to increase until the interactive mode ends, and after the interactive mode ends, the fan motor speed Raise.

(Example of washing machine)
When the home appliance 10 is a washing machine, the following processing is performed. During the interactive mode, the washing operation is not started even when the timer operation start time is reached (the operation mode is not entered), and the washing operation is started after the interactive mode ends.

(Heating cooker)
When the home appliance 10 is a heating cooker, the following processing is performed. During the interactive mode, the number of rotations of the fan that cools the inside of the product after the end of heating is decreased, and the rotation of the fan is restored after the interactive mode is completed.

(Cleaning robot)
When the home appliance 10 is a cleaning robot, the following processing is performed. During the interactive mode, the cleaning operation mode is stopped, and the operation is resumed after the interactive mode ends.

  As described in the specific example above, the home appliance 10 suppresses (low noise operation) or stops the operation of the operation unit 15 of the home appliance 10 in the interactive mode. In addition, the transition to the operation mode with a loud operation sound is limited until the interactive mode ends. If there is a performance or usability problem when restricting the operation mode until the end of the interactive mode, a message prompting the end of the interactive mode may be output and displayed to allow the user to be in the middle of the dialog with the user. You can exit interactive mode.

[Embodiment 2]
The following will describe another embodiment of the present invention with reference to FIGS. This embodiment demonstrates the case where the speech recognition system which concerns on this invention is applied to the household appliance network system with which a household appliance and a household appliance communicate. For convenience of explanation, members having the same functions as those described in the first embodiment are denoted by the same reference numerals and description thereof is omitted.

(Home Appliance Network System)
FIG. 2 is a block diagram showing a configuration of the home appliance network system (voice recognition system) 70 of the present embodiment.

  The home appliance network system 70 includes a home appliance (first unit) 10a and a home appliance (second unit) 10b, which are capable of wireless communication, and are connected to a wireless local area network (LAN) that is a narrow area communication network. Are interconnected. The type of home appliance 10a is not limited as long as it is a device that accepts input speech and executes speech processing that recognizes speech. If the household appliance 10b is an apparatus which performs the operation | movement which produces a sound (operation sound), a kind will not be limited. Further, the home appliance 10a and the home appliance 10b may be different types of home appliances such as a refrigerator and an air conditioner, or both may be the same type of home appliance such as an air conditioner.

  As shown in FIG. 2, the home appliance 10a includes a voice input unit 11, a voice output unit 12, a control unit 13a, a data storage unit 14, an operation unit 15, a sensor unit 16, a display unit 17, and a communication unit 18. . The control part 13a is the structure which is not provided with the determination part 133 of the control part 13 of the household appliance 10 demonstrated in Embodiment 1. FIG. Thus, the home appliance 10a has the same configuration as the home appliance 10 except that the home appliance 10a does not include the determination unit 133 but includes the communication unit 18.

  The home appliance 10b includes a control unit 13b, a data storage unit 14, an operation unit 15, a sensor unit 16, a display unit 17, and a communication unit 18, as shown in FIG. The control part 13b is a structure which is not provided with the audio | voice processing part 131 of the control part 13 of the household appliance 10 demonstrated in Embodiment 1. FIG. Therefore, the home appliance 10b has the same configuration as the home appliance 10 except that it does not include the voice input unit 11, the voice output unit 12, and the voice processing unit 131 but includes the communication unit 18.

  The communication unit 18 is a block that performs communication with the outside. When the home appliance 10a performs voice processing, the communication unit 18 of the home appliance 10a transmits information on the start of the voice processing to the home appliance 10b. When there are a plurality of home appliances 10b, they are transmitted to each. When the communication unit 18 of the home appliance 10b receives the information on the start of voice processing from the home appliance 10a, the determination unit 133 of the home appliance 10b is based on at least one of the sensing result of the sensor unit 16 of the home appliance 10b or the mode information of the home appliance 10b. The presence / absence of urgency of the operation of the operation unit 15 of the home appliance 10b is determined. And operation control part 132 of household appliance 10b controls operation of operation part 15 of household appliance 10b which is an own device. Since this control is similar to the control of the operation of the operation unit 15 of the home appliance 10 according to the first embodiment, the description thereof is omitted.

  Here, the home appliance 10b may control the device itself when the registered information on the installation location is the same as that of the home appliance 10a. The installation location information may be not only an address where the home appliance 10b is installed, but also information specifying an area such as a living room or a dining area. Therefore, the communication unit 18 of the home appliance 10a transmits information on the installation location of the own device to the home appliance 10b together with the information on the start of the voice processing. And the household appliance 10b controls the said own apparatus, when the installation place of the received household appliance 10a is the same as its own apparatus.

  In the home appliance network system 70, the home appliance 10a and the home appliance 10b are connected to the network, so that the home appliance 10b determines whether or not the operation of the home appliance 10b is urgent when the home appliance 10a is notified of the start of voice processing. The operation of the home appliance 10b can be controlled. In this way, it is possible to control the operation of the operation unit 15 of the home appliance 10b different from the home appliance 10a that performs voice processing. Since the household appliance 10a and the household appliance 10b are installed in the same installation place, it can suppress that the operation sound of the operation part 15 of the household appliance 10b enters into the input voice to the household appliance 10a as noise, and the voice recognition rate of the household appliance 10a Can be raised.

  As described above, the home appliance network system 70 controls the operation of the home appliance 10b different from the home appliance 10a that performs voice processing. The control of the operation of the home appliance 10b is the control of the output of the uttered voice of the home appliance 10b. Also good. This control will be described using a specific example. For example, when an air conditioner installed in the same room as the refrigerator automatically outputs a voice when it detects a start instruction to accept input voice in the refrigerator (start trigger occurs) and enters the interactive mode, the contents are urgent If there is no character (for example, chat), the air conditioner cancels (stops) the audio output. Also, when the start trigger occurs in the refrigerator and the conversation mode is entered, if the start trigger occurs even in an air conditioner installed in the same room as the refrigerator, the voice processing in the air conditioner is suppressed, and the message “Refrigerator is talking. If you say "yo" or "let's talk later" by outputting or displaying a sound with a small sound, the user can be notified of the reason why the user cannot interact with the air conditioner. Notification of such information may be performed by a remote controller of the air conditioner or a smartphone that can communicate with the air conditioner. Thus, when a start trigger occurs in a plurality of home appliances, it is preferable that exclusive control between home appliances is performed.

  Here, in the present embodiment, the home appliance 10a is a device that executes voice processing, and the home appliance 10b is a device that performs an operation that generates sound (operation sound). The configuration has been described as not including the unit 11, the audio output unit 12, and the audio processing unit 131. However, you may of course have it. Moreover, in this case, when the home appliance 10a includes the determination unit 133 and detects the start of the voice processing of the home appliance 10b, the home appliance 10a has a detection result based on at least one of the sensing result of the sensor unit 16 of the home appliance 10a or the mode information of the home appliance 10a. The structure which determines the presence or absence of the urgency of the operation | movement of the operation | movement part 15 may be sufficient. In other words, mutually (exclusively), when the start of voice processing of the communication partner's home appliance is detected, the presence or absence of the urgency of the operation of the sound of the own device is determined. The structure which performs control which suppresses or stops the operation | movement which the sound of an own device produces for the predetermined period including the reception period of an input audio | voice may be sufficient.

  In the present embodiment, only two home appliances are illustrated in FIG. 2 for the sake of simplicity. However, the number of home appliances included in the home appliance network system 70 is not limited. As shown in FIG. 3, there may be six home appliances included in the home appliance network system 70. In this case, among a plurality of home appliances, for example, the refrigerator 10-1, the air conditioner 10-2, the air purifier 10-3, the washing machine 10-4, the heating cooker 10-5, and the cleaning robot 10-6, Even the household appliance 10a which performs audio | voice processing may be the household appliance 10b which performs the operation | movement which produces an operation sound. In addition, some or all of a plurality of home appliances may be home appliances that perform sound processing and perform an operation that generates an operation sound.

[Embodiment 3]
Another embodiment of the present invention will be described below with reference to FIGS. This embodiment demonstrates the case where the speech recognition system which concerns on this invention is applied to the household appliance network system with which a household appliance and a management server apparatus communicate. For convenience of explanation, members having the same functions as those described in the first or second embodiment are denoted by the same reference numerals and description thereof is omitted.

(Home Appliance Network System)
FIG. 4 is a block diagram illustrating a configuration of the home appliance network system (voice recognition system) 100 according to the present embodiment.

  As shown in FIG. 4, the home appliance network system 100 includes a home appliance (first unit) 10A, a home appliance (second unit) 10B, and a management server device (server device) 20, which are the home appliance 10A and the home appliance 10B. Is connected to the management server device 20 by the wide area communication network 60.

  The home appliance 10 </ b> A is a device that receives input speech and executes speech processing that recognizes speech, and the type is not limited as long as it is a device connected to the wide area communication network 60. The home appliance 10 </ b> B is a device that performs an operation that generates sound (operation sound), and the type of the home appliance 10 </ b> B is not limited as long as it is a device connected to the wide area communication network 60. Further, the home appliance 10A and the home appliance 10B may be different types of home appliances such as a refrigerator and an air conditioner, or both may be the same type of home appliance such as an air conditioner. The home appliance 10 </ b> A and the home appliance 10 </ b> B are connected to the wide area communication network 60 and configured to be controllable via the wide area communication network 60. The home appliance 10 may be configured to connect a communication adapter having a communication function to a home appliance main body that does not have a connection function to the wide area communication network 60.

  The management server device 20 is a device that manages and controls the home appliances 10A and 10B via the wide area communication network 60. The management server device 20 may be configured such that a plurality of devices share functions instead of one. Further, as the management server device 20, for example, a cloud server that provides a cloud service may be used, but the management server device 20 is not limited to the cloud server.

  In the present embodiment, in order to simplify the description, a case where two home appliances 10A and 10B are connected to the management server device 20 will be described as shown in FIG. However, the number of home appliances included in the home appliance network system 100 is not limited. As shown in FIG. 3, there may be three or more. In this case, among a plurality of home appliances, for example, the refrigerator 10-1, the air conditioner 10-2, the air purifier 10-3, the washing machine 10-4, the heating cooker 10-5, and the cleaning robot 10-6, Even the home appliance 10 </ b> A that performs voice processing and is connected to the wide area communication network 60 may be the home appliance 10 </ b> B that performs an operation that generates an operation sound and is connected to the wide area communication network 60. In addition, some or all of the plurality of home appliances may be home appliances that perform voice processing, perform an operation that generates an operation sound, and are connected to the wide area communication network 60.

  For example, a wireless LAN is provided in the user home 50 where the home appliance 10A and the home appliance 10B are installed. The wireless LAN relay station 40 is connected to a wide area communication network 60 including the Internet. The relay station 40 is a communication device such as a WiFi (registered trademark) router or a WiFi access point. In the present embodiment, a configuration including the Internet is exemplified as the wide area communication network 60, but a telephone line network, a mobile communication network, a CATV (CAble TeleVision) communication network, a satellite communication network, or the like can also be used. And the household appliance 10 installed in the user's house 50 and the management server apparatus 20 can communicate via the wide area communication network 60 and the relay station 40 of wireless LAN.

  Here, although only one user home 50 is disclosed in FIG. 3, there is no limit on the number of user homes in which home appliances managed by the management server device 20 are installed.

  Home appliance 10A has the same configuration as home appliance 10a described in the second embodiment, as shown in FIG. As shown in FIG. 4, home appliance 10B has the same configuration as home appliance 10b except that it does not include determination unit 133 of control unit 13b of home appliance 10b described in the second embodiment.

  As shown in FIG. 4, the management server device 20 includes a control unit 21 that controls the operation of each unit of the management server device 20, a data storage unit 22 that stores various information used in the management server device 20, and an external device. A communication unit 23 that performs communication is provided. In addition, the control unit 21 functions as a determination unit 133 and a home appliance control unit (operation control unit) 212. The determination unit 133 has the same configuration as the determination unit 133 described in the first embodiment. Based on at least one of the sensing result of the sensor unit 16 of the home appliance 10B or the mode information of the home appliance 10B, the determination unit 133 determines whether or not the operation unit 15 of the home appliance 10B has an urgency. The home appliance control unit 212 is a block for remotely operating the home appliance 10B. The home appliance control unit 212 controls the operation of the operation unit 15 of the home appliance 10B via the communication unit 23.

  The data storage unit 22 of the management server device 20 registers and manages the mode information (operation state), the sensing result of the sensor unit 16, the installation location, and the like for the home appliance 10A and the home appliance 10B, respectively.

  When the home appliance 10A performs voice processing, the communication unit 18 of the home appliance 10A transmits information on the installation location of the own device to the management server device 20 together with information on the start of the voice processing. When the management server device 20 receives information on the installation location of the home appliance 10A from the home appliance 10A together with the information on the start of audio processing, the management server device 20 performs the following control on the home appliance 10B having the same location information as the home appliance 10A. The management server device 20 determines the presence / absence of urgency of the operation of the operation unit 15 of the home appliance 10B based on at least one of the sensing result of the sensor unit 16 of the home appliance 10B being managed and the mode information of the home appliance 10b. And the household appliance control part 212 controls the household appliance 10b via the communication part 23. FIG. Based on the control from the home appliance control unit 212, the operation control unit 132 of the home appliance 10B controls the operation unit 15 of the home appliance 10B. Since this control is similar to the control of the operation of the operation unit 15 of the home appliance 10 according to the first embodiment, the description thereof is omitted.

  As described above, since the home appliance network system 100 has the home appliances 10A and 10B and the management server device 20 connected to the network, the management server device 20 receives the notification of the start of voice processing from the home appliance 10A. The operation of home appliance 10B can be controlled by determining whether or not the operation of 10B is urgent. In this way, the management server device 20 can control the operation of the home appliance 10b different from the home appliance 10A that performs voice processing. Since the home appliance 10A and the home appliance 10B are installed at the same installation location, it is possible to suppress the operation sound of the operation unit 15 of the home appliance 10B from entering the input voice to the home appliance 10A as noise, and the voice recognition rate of the home appliance 10A Can be raised.

  Furthermore, the home appliance 10B includes a voice input unit 11, a voice output unit 12, and a voice processing unit 131, the determination unit 133 performs the above determination on the home appliance 10A, and the home appliance control unit 212 operates the operation unit 15 of the home appliance 10A. You may control operation | movement of. In this case, when the management server device 20 detects the start of the audio processing of the home appliance 10B, the operation of the operation unit 15 of the home appliance 10A is based on at least one of the sensing result of the sensor unit 16 of the home appliance 10A or the mode information of the home appliance 10a. The presence or absence of urgency is determined, and the operation of the operation unit 15 of the home appliance 10A is controlled. That is, when the management server device 20 detects the start of audio processing of any home appliance, the management server device 20 determines whether or not there is an urgency in the operation of generating sound from other home appliances, and a sound is generated for a predetermined period in a non-urgent home appliance. The structure which performs control which suppresses or stops operation | movement may be sufficient.

  As described above, in the present embodiment, the home appliance 10A and the home appliance 10B are connected to the management server device 20 via the wide area communication network 60, whereby the control integrated in the management server device 20 can be performed.

  Furthermore, as illustrated in FIG. 3, the home appliance network system 100 includes a communication terminal device 30 such as a tablet terminal or a smartphone, and the communication terminal device 30 can check the operation state of the home appliances 10A and 10B. There may be. In this case, the management server device 20 associates the communication terminal device 30 with the home appliance 10A and the home appliance 10B, and registers and manages the combination. By registering the combination in the management server device 20, the communication terminal device 30 is used to remotely operate the associated home appliance 10 or to confirm the operation state of the remotely associated home appliance 10. be able to. It is included in home appliance network system 100. The type and number of communication terminal devices 30 are not limited. Therefore, the management server device 20 can associate a plurality of home appliances as remote operation targets with respect to one communication terminal device 30. Moreover, it is possible to associate a plurality of communication terminal devices 30 as remote operation sources for one home appliance. The management server device 20 may notify the communication terminal device 30 of information indicating that the operation of the sound performed by the operation unit 15 of the home appliance 10B is suppressed or stopped as the operation state of the home appliance 10B. Although only one communication terminal device 30 is shown in FIG. 3, it is possible to notify a plurality of communication terminal devices associated with the home appliance B.

  The communication terminal device 30 should just be comprised so that a dedicated application can be downloaded and it can communicate with the household appliance 10 via the management server apparatus 20, for example. The communication terminal device 30 and the Internet in the wide area communication network 60 are connected using 3G (3rd Generation), LTE (Long Term Evolution), a home or public WiFi access point, or the like. Further, the communication terminal device 30 can communicate with the home appliances A and B via the relay station 40 without using the wide area communication network 60.

  The present invention is not limited to the above-described embodiments, and various modifications can be made. The embodiments of the present invention can also be obtained by appropriately combining technical means disclosed in different embodiments. Included in the range. Furthermore, a new technical feature can be formed by combining the technical means disclosed in each embodiment.

[Summary]
The speech recognition system (home appliance 10, home appliance network system 70, home appliance network system 100) according to aspect 1 of the present invention is a first unit (home appliance 10a, home appliance) that performs input voice reception and voice recognition of the received input voice. 10A), and a second unit (home appliance 10b, home appliance 10B) that performs a sound-generating operation, when a start instruction for accepting the input voice is detected, a sensor included in the second unit Based on at least one of the sensing results of the second unit and the control state of the second unit, a determination unit (133) that determines whether the operation is urgent or not, and the input voice when the determination unit determines that there is no urgency Operation control unit (132, home appliance control unit 212) that suppresses or stops the operation for a predetermined period including a reception period of , Comprising a.

  According to the above configuration, by determining whether or not there is an urgency of the operation of generating the sound of the second unit based on at least one of the sensing result of the sensor of the second unit or the mode information of the second unit, It is possible to perform processing in consideration of the harmful effects caused by the stoppage. That is, it is possible to prioritize the reception of the input voice when the operation that generates the sound is not urgent and to perform the process that prioritizes the operation when the operation that generates the sound is urgent.

  Furthermore, when it is determined that the operation that generates the sound is not urgent, the operation that generates the sound is suppressed or stopped for a predetermined period including the reception period of the input sound. In this case, it is possible to suppress the influence of the sound generated by the operation on the reception of the input sound. That is, since the sound by the second unit that enters the input speech as noise can be suppressed, the speech recognition rate can be increased. In addition, when there is no urgency, since the operation of generating sound by the second unit is suppressed or stopped, adverse effects due to the suppression or stop of this operation can be reduced.

  As can be seen from the above, according to one embodiment of the present invention, it is possible to provide a speech recognition system that can increase the speech recognition rate and reduce adverse effects caused by suppression or stop of the operation that generates sound. For example, when the first unit is a refrigerator and the second unit is an air conditioner, adverse effects caused by the suppression or stoppage of the sound-generating operation include heat stroke of the user due to the suppression or stoppage of the operation of the compressor of the air conditioner. It is.

  The speech recognition system (home appliance 10) according to aspect 2 of the present invention is the speech recognition system according to aspect 1, in which the first unit, the second unit, the determination unit, and the operation control unit are integrated devices. (Home appliance 10).

  According to the above configuration, when an instruction to start receiving input voice is detected in one device, the urgency of the operation of the device's sound is determined. Therefore, in one device, it is possible to select which of the operation that generates a sound, the reception of the input speech, and the speech recognition is prioritized. Therefore, according to the above configuration, it is possible to provide a device capable of increasing the voice recognition rate of the own device while reducing the adverse effects caused by the stop of the operation that generates the sound of the own device. That is, it is possible to provide a device that is easy to use for the user and that can perform voice recognition accurately. Such a device may be, for example, a home electronic device, a communication robot, or the like. For example, in the case where the integrated device is a refrigerator, the harmful effects caused by the suppression and stoppage of the sound-generating operation include suppression of the operation of the compressor of the refrigerator and an increase in the internal temperature due to the stoppage.

  The speech recognition system (home appliance network system 70) according to aspect 3 of the present invention is the speech recognition system according to aspect 1, wherein the second unit (home appliance 10b) includes the determination unit and the operation control unit. 1 unit (home appliance 10a) and the second unit are connected via a communication network, and the determination unit receives information indicating that the input voice reception start instruction has been detected from the first unit. Then, the above determination is performed.

  According to the above configuration, the first unit and the second unit are connected via the communication network, and the operation of generating the sound of the second unit different from the first unit that accepts the input voice and recognizes the voice is controlled. It can be carried out. Here, when the second unit receives the information on the installation location of the first unit together with the information on the reception start of the input voice from the first unit, the information on the installation location of the second unit is the same as that of the first unit. In this case, it is sufficient to control itself. The installation location information may be not only an address where the unit is installed, but also information specifying an area such as a living room or a dining area. Both the first unit and the second unit may be electronic devices connected via a communication network.

  The speech recognition system (home appliance network system 100) according to aspect 4 of the present invention is the speech recognition system according to aspect 1, in which the first unit (home appliance 10A), the second unit (home appliance 10B), and the determination. And a server device (management server device 20) including the operation control unit are connected via a communication network, and the determination unit detects an instruction to start receiving the input voice from the first unit. When the information indicating that it has been received is received, the above determination is made.

  According to the above configuration, since the first unit, the second unit, and the server device are connected via the communication network, it is possible to perform control integrated in the server device. It is possible to control the operation of generating a sound in a second unit different from the first unit that receives input speech and recognizes speech. When the second unit is around the first unit, for example, at the same installation location (for example, the same address, living room, kitchen), the effect of increasing the voice recognition rate of the first unit is exhibited. The first unit and the second unit may be electronic devices that are connected to the server device via a communication network.

  In the speech recognition system according to aspect 5 of the present invention, in the speech recognition system according to any one of aspects 1 to 4, the determination unit determines the urgency a plurality of times during the predetermined period, and the determination unit When it is determined that there is urgency, the operation control unit executes the operation.

  According to the above configuration, the operation is executed when the operation of generating the sound of the second unit is urgent. Therefore, no adverse effects are caused by suppressing or stopping the operation that generates sound.

  In the speech recognition system according to Aspect 6 of the present invention, in the speech recognition system according to any one of Aspects 1 to 5, the determination unit includes a sensing result of a sensor included in the second unit or a detection result of the second unit. A value indicating urgency is calculated based on at least one of the control states. When the value is equal to or less than a predetermined value, it is determined that there is no urgency, and when the value exceeds a predetermined value, it is determined that there is urgency.

  According to the said structure, the presence or absence of urgency can be determined by the value which shows the urgency calculated based on at least one of the sensing result of the sensor of a 2nd unit, or the control state of a 2nd unit. In this case, for example, the higher the value indicating the calculated urgency, the higher the value of the calculated urgency, the normal operation (when there is no voice reception and voice recognition in the first unit), and the lower the value, the higher the operation. You may perform control which suppresses more.

  In the speech recognition system according to aspect 7 of the present invention, in the speech recognition system according to any one of aspects 1 to 6, the first unit transmits the received input speech to an external device that recognizes speech, The voice recognition is executed by receiving a voice recognition result from the external device as a response to the transmission.

  According to the above configuration, since the external device performs voice recognition, the cost of the first unit can be suppressed and the voice recognition function can be improved without increasing the processing capacity of the first unit or expanding the storage capacity. it can. In addition, when the external device has high performance, more information can be provided to the user as a voice recognition result than when the external device is not connected. Moreover, if the external device is configured to perform voice recognition of the plurality of first units, the cost can be suppressed. Here, the external device for voice recognition may be the server device according to the fourth aspect.

  In the speech recognition system according to aspect 8 of the present invention, in the speech recognition system according to any one of aspects 1 to 7, the sound is an operation sound caused by the operation, or an operation that generates the sound. Is the output of the speech.

  According to the above configuration, if the operation of generating the sound of the second unit is not urgent, this operation is suppressed, or if the utterance of the second unit is not urgent, the output of the utterance is suppressed or stopped. By doing so, the speech recognition rate of the first unit can be increased.

  In addition, when the second unit also receives input speech and performs speech recognition, it determines the urgency of input speech and speech recognition of both the first unit and the second unit, and performs exclusive control between the units. May be. In this case, it may be determined that a unit that has started receiving voice input later in time sequence has no urgency. Alternatively, voice input priority is set between units, and a unit with a low priority may be determined to have no urgency.

  In the speech recognition system according to aspect 9 of the present invention, in the speech recognition system according to any one of aspects 1 to 8, the input after the first unit detects an instruction to start accepting the input speech. This is a period until a voice reception end instruction is detected.

  According to the above configuration, the period determined by the detection of the input voice reception start instruction and the detection of the input voice reception end instruction can be a period during which the operation of generating the sound of the second unit is suppressed or stopped. . The input voice reception start instruction and the input voice reception end instruction are not limited as long as the information can be detected by the first unit.

  The input voice reception start instruction is, for example, by pressing down a button for starting input voice provided in the first unit, a specific voice (for example, “Speak! A configuration in which the first unit can be detected by input of a sound such as tapping, etc. may be used. If the first unit is a device that interacts with the user, the end instruction for accepting the input sound is that there is no input sound for a certain period of time after the end of the conversation is detected by the content of the conversation or the utterance sound is output from the first unit The structure which a 1st unit detects by detection may be sufficient. Also, the end instruction is given by pressing down an input voice end button (for example, forced end button) provided in the first unit, a specific voice (for example, a user voice of “seeking”), or two consecutive times. The first unit may be able to detect the sound by hitting a hand or the like.

  The speech recognition system according to aspect 10 of the present invention is the speech recognition system according to any one of aspects 1 to 9, wherein when the operation control unit suppresses or stops the operation, the operation is suppressed or stopped. A notification unit that notifies the information to be indicated.

  According to the above configuration, since the notification unit notifies the information that the operation of generating the sound of the second unit is suppressed or stopped, the user can know the information by the notified information.

  The notification may be, for example, a notification by display on the display unit provided in the second unit or a voice notification from the audio output unit. Alternatively, the notification may be a notification to an external communication terminal device.

  A control method for a speech recognition system according to an aspect 11 of the present invention includes a first unit that performs reception of input speech and speech recognition of the received input speech, and a second unit that performs operations that generate sound. In the control method of the voice recognition system, when the start instruction to accept the input voice is detected, the urgency of the operation is based on at least one of the sensing result of the sensor of the second unit and the control state of the second unit. A determination step for determining whether or not there is an emergency, and an operation control step for suppressing or stopping the operation for a predetermined period including the reception period of the input voice when it is determined that there is no urgency in the determination step.

  According to the above control method, the voice recognition system is controlled so that the same effects as the voice recognition system can be obtained, and the adverse effects caused by the suppression or stop of the operation that generates sound can be reduced while increasing the voice recognition rate. be able to.

  The speech recognition system according to each aspect of the present invention may be realized by a computer. In this case, by causing the computer to operate as a determination unit and an operation control unit included in the voice recognition system, a program for realizing the voice recognition system on the computer, and a computer-readable recording medium that records the program are also provided. It falls within the scope of the present invention.

  The present invention can be used for a speech recognition system having a speech recognition function, for example, home appliances, robots, and the like.

10 Home appliances (voice recognition system, first unit, second unit)
10a Consumer electronics (first unit)
10b Home appliances (second unit)
DESCRIPTION OF SYMBOLS 11 Audio | voice input part 12 Audio | voice output part 13, 13a, 13b Control part 14 Data storage part 15 Operation | movement part 16 Sensor part (sensor)
17 Display unit 18 Communication unit 20 Management server device (server device)
60 Communication network 70,100 Home appliance network system (voice recognition system)
131 Voice processing unit 132 Operation control unit 133 Determination unit 212 Home appliance control unit (operation control unit)

Claims (4)

In a speech recognition system comprising: a first unit that performs reception of input speech and speech recognition of the received input speech; and a second unit that performs operations that generate sound.
When detecting an instruction to start accepting the input voice, a determination unit that determines the presence or absence of urgency of the operation based on at least one of a sensing result of the sensor of the second unit or a control state of the second unit;
When the determination unit determines that there is no urgency, the operation is suppressed or stopped for a predetermined period including the reception period of the input voice,
When the determination unit determines that there is urgency, the operation control unit executes the operation and outputs or displays a message explaining the reason to the user in the first unit or the second unit. Preparation
The determination unit calculates a value indicating urgency based on at least one of a sensing result of the sensor of the second unit or a control state of the second unit, and if the value is equal to or less than a predetermined value, there is no urgency If the value exceeds the predetermined value, it is determined that there is urgency,
The speech recognition system according to claim 1, wherein when the determination unit determines that there is urgency, the content of the operation executed by the operation control unit varies depending on a value indicating the urgency . The second unit includes the determination unit and the operation control unit,
The first unit and the second unit are connected via a communication network;
The speech recognition system according to claim 1, wherein the determination unit performs the determination when receiving information indicating that an instruction to start accepting the input speech has been detected from the first unit. The first unit;
The second unit;
A server device including the determination unit and the operation control unit is connected via a communication network,
The speech recognition system according to claim 1, wherein the determination unit performs the determination when receiving information indicating that an instruction to start accepting the input speech has been detected from the first unit.   The voice recognition system according to claim 1, wherein the determination unit determines the urgency a plurality of times during the predetermined period.

Images