JP2007121579A - Operation device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an operation device capable of operating equipment in response to only a voice that a user intentionally gives to operate the equipment. <P>SOLUTION: A voice recognition means 3 judges whether the meaning of a voice input to a microphone 2 matches a preregistered instruction and outputs the instruction to an operation means 4 when the meaning of the voice matches the instruction. A trigger means 10 outputs a trigger signal to the operation means 4 when detecting a prescribed sign from the user. The operation means 4 operates an air conditioner A or a television TV as an object to be operated according to the instruction from the voice recognition means 3 only in a prescribed acceptance period after the trigger signal is received. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an operating device that operates a device to be operated according to an input voice.

  Conventionally, as this type of operation device, there is a voice input means (microphone) for converting voice into a voice signal composed of an electrical signal, and whether there is a command that matches the meaning of the voice among pre-registered commands. A speech recognition means (speech recognition processing circuit) for determining the command based on the speech signal, and a controller that generates a command signal corresponding to the command when the speech recognition means determines that the command matches the meaning of the speech. What you have is provided. The generated command signal is transmitted from the transmission device to a device to be operated, and the device is operated according to the command signal (see, for example, Patent Document 1).

In the operation device described in Patent Literature 1, the operation target is an air conditioner, and the air conditioner is operated so that the blowing becomes stronger when, for example, a voice of “strong” is input. As a result, it is possible to operate the device with only sound without requiring manual operation by the user.
JP 7-253237 A (page 3)

  However, in the operation device described in Patent Document 1, even if the sound is other than the sound generated when the user intends to operate the device, the device is operated when the meaning of the sound matches the command. Therefore, if a voice (for example, “strongly”) having the same meaning as the command is included in the conversation between people, the device may be erroneously operated in response to the voice. That is, an operation unintended by the user may be performed by reacting to a sound other than the sound generated when the user intends to operate the device.

  The present invention has been made in view of the above reasons, and an object of the present invention is to provide an operating device capable of operating a device in response to only a sound generated by a user intending to operate the device.

  According to the first aspect of the present invention, there is provided a voice input means for converting voice into a voice signal that is an electrical signal, and whether or not the meaning of the voice inputted to the voice input means matches a pre-registered command. A voice recognition unit that is determined based on the command, an operation unit that operates the device to be operated in accordance with the command when the voice recognition unit determines that the meaning of the voice matches the command, and a predetermined value from the user And a trigger unit that outputs a trigger signal when the signal is detected by the detection unit, and is input to the voice input unit during a predetermined reception period after the trigger signal is output. The device is operated only when the meaning of the voice matches the command.

  According to this configuration, the operating device operates the device only when the meaning of the voice input to the voice input unit in the reception period matches the command. Therefore, any voice is input during the period other than the reception period. Even if it is input to the means, the device is not operated. On the other hand, since the reception period is set after the trigger means detects a predetermined signal from the user by the detection means and outputs a trigger signal, the reception period can be started by the user sending a signal, In other words, the acceptance period will not start unless the user sends a signal. In other words, the invention of claim 1 is such that the device can be operated by voice only after the user sends a signal, and therefore the user is not intended to operate the device and the signal is sent. In such a state, even if, for example, a voice whose meaning matches the command is included in a conversation between people, the device is not erroneously operated in response to the voice. Therefore, there is an advantage that the device can be operated only in response to the sound generated by the user intending to operate the device. In addition, since the reception period is set after detecting a signal from the user, once the signal from the user is detected and the reception period starts, the user sends a new signal within this reception period. The device can be operated by voice. That is, the procedure for operating the device is simple.

  According to a second aspect of the present invention, in the first aspect of the invention, the detecting means includes a line-of-sight detection unit that detects the line of sight of the user, and the cue is that the line of sight of the user is directed in a specific direction. It is characterized by.

  According to this configuration, since the user can operate the device by voice only by directing his / her line of sight in the specific direction, the user can operate the device without moving the limbs.

  The invention of claim 3 is characterized in that, in the invention of claim 1, the detection means includes a microphone for detecting sound, and the signal is a specific sound.

  According to this configuration, since the user can operate the device only by generating the specific sound, the device can be easily operated even when operating the device in a dark room or when the user is visually impaired. Can be operated.

  According to a fourth aspect of the present invention, in the first aspect of the invention, the detection unit includes a camera that captures the user, and an image processing unit that detects the gesture of the user from the image captured by the camera. The cue is a specific gesture of the user.

  According to this configuration, since the user can operate the device only by performing the specific gesture, the device can be easily operated even when operating the device in a dark room or when the user is visually impaired. Can be operated. Gesture here refers to, for example, a so-called peace sign in which the user opens the forefinger and middle finger of one hand in a V shape and bends the other fingers, or creates a ring with the thumb and forefinger of one hand and other fingers. It means that specific signs such as spread so-called okay signs are created and these signs are directed to the camera, and the user takes a specific pose using the whole body.

  The invention of claim 5 is the invention according to any one of claims 1 to 4, wherein the trigger means has a plurality of the detection means, and the trigger is detected when the detection signal is detected by all the detection means. A signal is output.

  According to this configuration, since the reception period starts and the device can be operated by voice only after the signal is detected in all of the plurality of detection means, it is assumed that the signal is erroneously detected by one detection means. However, as long as the signal is not detected by the remaining detection means, the device cannot be operated by voice. Therefore, it is possible to prevent the reception period from being erroneously started in a state where the user does not intend to operate the device, and it is possible to reliably avoid erroneous operation due to sound other than the sound generated by the user intending to operate the device. There are advantages.

  According to a sixth aspect of the present invention, in any one of the first to fifth aspects, the operation means transmits the command to the device using a radio signal.

  According to this configuration, since it is not necessary to route an electric wire between the operating device and the device to be operated, wiring work at the time of installing the operating device or the device is simplified. The radio signal here includes, for example, a signal transmitted by infrared communication, a signal transmitted using radio waves as a carrier wave, and the like.

  According to the present invention, since the trigger means detects the signal from the user by the detection means, the device can be operated by voice only. Therefore, the user does not intend to operate the device. If no voice is sent, for example, even if a voice whose meaning matches the command is included in a conversation between people, the device will not be operated by mistake in response to the voice. Therefore, there is an advantage that the device can be operated only in response to the sound generated by the user intending to operate the device.

  The operating device according to the present embodiment operates a device as an operation target in accordance with a voice generated by a user. As shown in FIG. 2, the user U lies on a bed B installed indoors. The apparatus main body 1 is provided on the wall surface near the head side of the bed B so that the device can be operated in a lying posture. In the present embodiment, an example is shown in which an air conditioner (air conditioner) A and a television (television receiver) TV installed in the same room as the apparatus body 1 are operated.

  In the following, first, a configuration for operating the air conditioner A and the television TV, which are operation targets, according to the input voice will be described with reference to FIG.

  The apparatus main body 1 includes a microphone 2 as sound input means for converting sound into a sound signal composed of an electric signal and outputting the sound signal, and a control circuit for receiving the sound signal from the microphone 2. The microphone 2 is installed at a position where the sound generated by the user U lying on his / her back on the bed B can easily reach. The control circuit is provided with a voice recognition unit 3 that performs voice recognition based on a voice signal, and an operation unit 4 that operates the air conditioner A and the television TV according to the output of the voice recognition unit 3.

  The speech recognition means 3 includes an acoustic model storage unit 5 in which data for extracting feature parameters of an input speech including phonemes and prosody from a speech signal is stored in advance, and words included in the input speech based on the extracted feature parameters The language model / dictionary storage unit 6 in which data for extracting the voice is stored in advance, the acoustic model storage unit 5 and the language model / dictionary storage unit 6 are used to understand the meaning of the input speech, and the meaning of the input speech is registered in advance. And a voice recognition unit 7 that determines whether or not the command matches the command. Here, a plurality of types of commands used for operating the air conditioner A and the television TV such as “power on” and “power off” are registered.

  Here, if there is a command that matches the meaning of the input voice, the voice recognition unit 7 outputs this command to the operation command unit 9 constituting the operation means 4 together with the transmission unit 8 to be described later. If there is no output, no output is performed. However, even if both of the meaning of the input speech and the command do not completely match, it is sufficient if the similarity is determined to match if the similarity exceeds the threshold. In consideration of the balance between misrecognition that outputs an incorrect command and unrecognition that does not output the command, it is desirable to set the operation device so as not to hinder the operation.

  The operation command unit 9 receives a command from the voice recognition unit 7 and generates an operation signal corresponding to this command. The operation signal generated in this way is output to the transmission unit 8 subsequent to the operation command unit 9, and is transmitted from the transmission unit 8 to the air conditioner A or the TV TV. The air conditioner A and the television TV operate according to the operation signal when receiving the operation signal from the transmission unit 8. The distinction between the operation of the air conditioner A and the operation of the TV TV is made by the voice recognition means 3 recognizing that the voice prefixed with the word “air conditioner” relates to the operation of the air conditioner A and leading the word “TV” This is performed by recognizing that the sound to which the is attached relates to the operation of the television TV. That is, for example, when a voice “Set air conditioner to 28 degrees” is input, the operating device performs an operation of setting the set temperature of air conditioner A to 28 degrees, while, for example, a voice “Turns down the volume of the TV”. Is input, the TV TV volume is reduced.

  In the present embodiment, the transmission unit 8 operates by infrared communication by adopting a configuration in which each of the transmission unit 8 transmits infrared rays to a light receiving unit (not shown) for an infrared remote controller provided in each of the air conditioner A and the TV TV. A signal is transmitted. That is, a command from the operating device is transmitted to the device using an operation signal that is a radio signal. Here, as each operation signal for operating the air conditioner A, the same signal as the remote control signal transmitted at the time of various operations is used from the dedicated infrared remote controller (not shown) of the air conditioner A, and the TV TV is operated. As each operation signal, the same signal as the remote control signal transmitted at the time of various operations from a dedicated infrared remote controller (not shown) of the TV TV is used.

  Although an example in which a command is transmitted using a wireless signal is shown here, the command may be transmitted in a wired manner by connecting the device to be operated and the transmission unit 8 through a wired connection. As an example, the transmitter 8 and the device can be wiredly connected through the wall, and in this case, the apparatus body 1 adopts a configuration in which a part is embedded in the wall in the same manner as a well-known embedded wiring device. You can also In addition, when a device (for example, a lighting fixture) operated by a switch installed in advance on a wall surface is set as an operation target, the apparatus main body 1 provided with a relay (not shown) that opens and closes according to a command from the voice recognition unit 7. It is possible to operate the device with the relay of the device body 1 simply by replacing the existing switch with the device body 1 and connecting the device body 1 and the device using the preceding wiring for the switch. become.

  By the way, in this embodiment, as described above, the period for operating the air conditioner A and the TV TV according to the sound input to the microphone 2 is limited to a predetermined reception period. In other words, during a period other than the reception period, even if a sound corresponding to any command is input to the microphone 2, the air conditioner A and the TV TV are not operated according to the sound. Here, the reception period is defined by the operation command unit 9, and the operation command unit 9 defines a predetermined period after receiving a trigger signal described later as the reception period.

  Here, in order to determine the timing for starting the reception period, there is provided trigger means 10 for detecting a predetermined signal from the user U and outputting a trigger signal to the operation command section 9 when the signal is detected. It has been. The trigger means 10 of this embodiment has a detection means for detecting the line of sight of the user U, and outputs a trigger signal when the detection means detects that the line of sight is directed in a specific direction. Here, the specific direction is a direction of a video camera 11 (described later) provided in the apparatus main body 1, and a trigger signal is output when the line of sight of the user U is directed to the video camera 11 of the apparatus main body 1. In FIG. 1, the detection means is provided with a function for outputting a trigger signal, and the trigger means 10 is constituted by only the detection means. However, the detection means and the means for outputting the trigger signal may be provided separately. .

  Specifically, the trigger means 10 is a video camera 11 provided in the apparatus main body 1 so that the face of the user U can be photographed, and an image for executing preprocessing on an image signal output from the video camera 11. The signal processing unit 12 and the visual line detection unit 15 that receives the output of the image signal processing unit 12 and detects whether or not the visual line of the user U is directed toward the video camera 11 are configured. For example, a technique for extracting the shape of the pupil of the user U and determining that the line of sight is directed toward the video camera 11 is applied to the line-of-sight detection unit 15. The image signal processing unit 12 and the line-of-sight detection unit 15 are provided in a control circuit.

  The line-of-sight detection unit 15 outputs a trigger signal when detecting that the line of sight is directed to the video camera 11 of the apparatus main body 1. However, the line-of-sight detection unit 15 does not output a trigger signal when the line of sight is directed to the video camera 11 for a moment, but when the line of sight is directed continuously to the video camera 11 for a predetermined time, that is, the video camera 11. Is configured to output a trigger signal when the camera is watched. The operation command unit 9 starts the reception period after receiving the trigger signal, and ends the reception period when a certain time has elapsed. The operation command unit 9 operates in a reception state in which the command from the voice recognition unit 7 is valid only during the reception period, and in a standby state in which the command from the voice recognition unit 7 is invalid in a period other than the reception period. Operate. That is, the operation command unit 9 generates an operation signal only when an instruction is received during the reception period. Here, the voice recognition unit 7 is configured to receive a signal from the operation command unit 9 and perform a voice recognition process only during the reception period.

  The specific direction may be the direction of the device to be operated (air conditioner A or TV TV), and a trigger signal may be output when the trigger means 10 detects that the line of sight of the user U is directed to the device. . In this configuration, the user U can operate the device while visually confirming the operation state of the device to be operated. For example, when the device is not turned on, the user U first turns on the device. This makes it easy to perform operations according to the operating state of the device. Further, in this configuration, since the user U can operate the device by gazing at the device to be operated, the device operation can be performed by gazing at the video camera 11 of the apparatus main body 1 that is not related to the device. Intuitive operation is possible compared to a configuration that enables this.

  Hereinafter, the operation of the operation device of the present embodiment will be briefly described with reference to the flowchart of FIG. 3 showing the operation of the operation means.

  First, the operation command unit 9 is set to a standby state (step S01), and continues to operate in the standby state during a period until a trigger signal is input (a period other than the reception period). Therefore, during this period, for example, even if a voice corresponding to the command “Set the air conditioner to 28 degrees” (“Set the air conditioner to 28 degrees”) is input to the microphone 2, the operating device operates the device. There is no. On the other hand, when the user U gazes at the video camera 11 and a line of sight is directed toward the video camera 11, a trigger signal from the trigger means 10 is input to the operation command section 9 (step S02: Y), and thereafter Then, the operation command unit 9 shifts to the reception state (step S03) and starts a timer period T that defines the reception period (step S04). During the reception period, for example, when a voice “Set the air conditioner to 28 degrees” is input to the microphone 2, an instruction “Set the air conditioner to 28 degrees” is issued as an instruction that matches the meaning of the voice. (Step S05: Y), an operation signal for operating the air conditioner A so as to set the set temperature to 28 degrees is generated by the operation command section 9 (step S08). The generated operation signal is transmitted from the transmission unit 8 to the air conditioner A (step S09), and the air conditioner A sets the set temperature to 28 degrees according to the received operation signal.

  Further, since the reception period is set while the timer time T is limited to the fixed time Tc, if no sound is input within the reception period, the operation command is given at the end of the reception period (step S07: Y). The unit 9 shifts to a standby state (step S01). However, when a command is output from the voice recognition unit 7 to the operation command unit 9 during the reception period, the operation command unit 9 ends the reception period when the command is received.

  According to the configuration described above, the user U can operate the air conditioner A and the TV TV by the operation device without moving the limbs while keeping the posture lying on the bed B. However, even if a sound corresponding to any command (for example, “turn on the TV”) is input to the microphone 2 during a period other than the reception period, the operation device is the device (air conditioner A and TV TV). No operation is performed, and even if, for example, a voice saying “Turn on the power of the television” during a conversation between people is input to the microphone 2 in a period other than the reception period, the operation device reacts to this voice and the device Will not be manipulated. Accordingly, the operating device operates the air conditioner A and the TV TV only in response to the sound generated when the user U intends to operate the air conditioner A and the TV TV, thereby avoiding an operation unintended by the user U. be able to.

  As shown in FIG. 1, a speaker 17 and a lamp 18 are connected to the operation command section 9 via an output control section 16 provided in the control circuit. In this embodiment, the speaker 17 and the lamp 18 and the output are output. The control unit 16 is used as a notification unit 19 that notifies the user U that it is a reception period. That is, when the acceptance period starts, the operation command unit 9 turns on the lamp 18 via the output control unit 16 and outputs a predetermined voice (for example, “Please give me a request”) from the speaker 17 by voice synthesis. Thus, the user U is notified by light and voice that it is a reception period. This can prompt the user U to generate sound. Furthermore, in the unlikely event that the reception period starts regardless of the intention of the user U, the user U should be warned not to generate sound that may lead to an erroneous operation of the air conditioner A or TV TV. Can do.

  By the way, the configuration of the trigger means 10 is not limited to the configuration in which the user U's line of sight is directed in a specific direction and outputs a trigger signal as described above. It is also possible to adopt the configuration described in (1).

  That is, the trigger means 10 is configured to include a microphone in the detection means and to output a trigger signal when a specific sound is input to the microphone. Although the whistling sound is a specific sound here, not only the whistling, but also a specific sound such as tongue hitting, applause, or a specified voice (for example, “voice operation”) may be used. However, since a reception period starts when a specific sound is input and the air conditioner A and the TV TV are operated according to the input sound, the prescribed sound avoids a sound frequently generated in daily conversation. desirable.

  In the case of a configuration in which a prescribed sound is detected and a trigger signal is output, the microphone 2 as the voice input means shown in FIG. 1 is used as the microphone of the detecting means, and the trigger signal is recognized when the prescribed voice is recognized. It is desirable to configure the voice recognition means 3 to output. Thereby, since the voice input means and the voice recognition means 3 can be used as the trigger means, the function as the trigger means can be realized without adding a dedicated configuration as the trigger means. However, in this case, it is necessary to configure the voice recognition means 3 so that the voice recognition process is performed during a period other than the reception period.

  Further, the trigger signal may be output by taking the logical product of the detection result of whether or not the line of sight is directed in a specific direction and the detection result of whether or not the specific sound is detected. Specifically, a logical product circuit (not shown) is provided in the front stage of the operation command unit 9 to detect whether the line of sight is directed in a specific direction and whether the specific sound is detected. The input of the logical product circuit is used, and the output of the logical product circuit is used as the trigger signal. In this configuration, since the trigger signal is output and the reception period starts only when the line of sight is directed in a specific direction and a specific sound is detected, the line of sight directed in the specific direction and the specific sound are started. Even if only one of these is detected, the acceptance period does not start. As a result, there is an effect that it is possible to avoid the start of the reception period against the intention of the user U.

  Alternatively, for example, the detection unit includes a video camera provided in the apparatus main body 1 so that the upper body of the user U can be photographed, and an image processing unit that detects the gesture of the user U from the image of the video camera. The trigger means 10 can also be configured to output a trigger signal when it detects that a specific gesture has been performed. Here, a specific gesture is that the user U points a so-called peace sign in which the index finger and middle finger of one hand are opened in a V shape and the other fingers are bent toward the apparatus main body 1. In the image processing unit, the contour shape of the hand of the user U when the user U performs a peace sign toward the video camera is registered in advance as a standard pattern, and the user U's hand photographed by the video camera is registered. If the contour shape of the image matches the standard pattern, it is determined that the user U is performing a peace sign. The specific gesture is not limited to a peace sign, but a user U in a posture lying on his / her back on the bed B, such as a so-called okay sign in which a ring is formed with the thumb and index finger of one hand and the other fingers are spread out. It is desirable to be able to do almost without moving the body.

It is a block diagram which shows the structure of embodiment of this invention. It is explanatory drawing which shows the example of installation same as the above. It is a flowchart which shows operation | movement of the operation means same as the above.

Explanation of symbols

2 Microphone (voice input means)
3 Voice recognition means 4 Operation means 10 Trigger means 15 Line-of-sight detection part A Air conditioner (device)
TV TV (equipment)
U user

Claims (6)

  Voice input means for converting voice into a voice signal that is an electrical signal, and voice recognition for determining whether the meaning of the voice input to the voice input means matches a pre-registered command based on the voice signal And means for operating a device to be operated in accordance with the instruction when the voice recognition means determines that the meaning of the voice matches the instruction, and a detecting means for detecting a predetermined signal from the user. Trigger means for outputting a trigger signal when the signal is detected by the detection means, and the meaning of the voice input to the voice input means during a predetermined reception period after the trigger signal is output An operating device that operates a device only when they match.   The operation device according to claim 1, wherein the detection unit includes a line-of-sight detection unit that detects the line of sight of the user, and the cue is that the line of sight of the user is directed in a specific direction.   2. The operating device according to claim 1, wherein the detecting means includes a microphone for detecting sound, and the signal is a specific sound.   The detection means includes a camera that captures a user and an image processing unit that detects a user's gesture from an image captured by the camera, and the signal is a specific gesture of the user. The operating device according to claim 1.   5. The trigger device according to claim 1, wherein the trigger device includes a plurality of the detection devices, and outputs the trigger signal when all the detection devices detect the signal. The operating device according to 1.   The operation device according to claim 1, wherein the operation unit transmits the command to the device using a radio signal.

Images