JP2010278702A - Operation controller and operation control method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To discriminate speech intended by a user and noise not intended by the user without increasing a cost and power consumption. <P>SOLUTION: A VOX circuit 1 includes a CPU 5 which divides the output signal of a microphone by prescribed time unit (4 [ms]), calculates the values of the output signals corresponding to the largeness of sound by prescribed time unit, based on the divided output signals, compares the value of each output signal with a predetermined threshold, comparing the value of the output signal with the value of the output signal of the last prescribed time unit when the value of the output signal exceeds the threshold, and then, and performs VOX transmission when the value of the output signal is not lower than the value of the output signal of the last prescribed time unit three times or more continuously and also when the value of the output signal exceeds the threshold five times or more continuously. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an operation control apparatus and an operation control method.

2. Description of the Related Art Conventionally, there is an operation control technique based on voice detection as one of power saving techniques in a device (for example, a wireless communication device or a voice recorder) that involves voice input. The technology of operation control based on the detection of sound is to detect the output signal by automatically turning on the power of the device or a part of the device when the input of the output signal performed through the sound input unit such as a microphone is detected. This is a technique for automatically turning off the power of a device or a part of the device when there is no sound or after a predetermined time has elapsed since the detection of the voice.
Among the operation controls based on the detection of voice, in particular, the operation control used in the wireless communication device is called VOX (Voice Operated Transmitter). Hereinafter, VOX will be described as an example of operation control based on voice detection.

FIG. 8 is a block diagram showing a main configuration of a conventional VOX circuit 100. As shown in FIG.
A VOX circuit 100 shown in FIG. 8 is a circuit for performing operation control based on voice detection in a wireless communication device. As shown in FIG. 8, the VOX circuit 100 includes a microphone 2 (MIC2 in FIG. 8), an amplifier 3 (AMP3 in FIG. 8), a detection smoothing circuit 4 (DET4 in FIG. 8), and a CPU 5 (Central Processing Unit). .

The microphone 2 converts the sound into an electric signal (hereinafter referred to as “output signal”) and outputs it. An output signal from the microphone 2 is input to the amplifier 3.
The amplifier 3 amplifies the output signal from the microphone 2 and outputs it. The amplified output signal is input to the detection smoothing circuit 4.
The detection smoothing circuit 4 detects the amplified output signal input from the amplifier circuit and smoothes the detected output signal, and outputs the smoothed output signal. The smoothed output signal is input to the CPU 5.

  The CPU 5 performs analog / digital conversion (hereinafter referred to as “A / D conversion”) of the smoothed output signal to obtain a digitized output signal. Then, it is determined whether or not the value of the digitized output signal exceeds a predetermined threshold value. Here, the value of the digitized output signal indicates the loudness of the sound. That is, the CPU 5 determines whether or not a sound having a level exceeding a level based on a predetermined threshold is detected.

When the value of the digitized output signal exceeds a predetermined threshold value, the CPU 5 performs VOX transmission. VOX transmission is a process of outputting a signal for turning on the power of a transmitter (not shown) of a wireless communication device. A signal for turning on the power of the transmitter is input to a configuration for switching the power on / off of the transmitter via a terminal TX shown in FIG. The transmitter that is turned on starts up. The activated transmitter enables the wireless communication device to transmit voice.
If the value of the digitized output signal does not exceed a predetermined threshold, the transmitter is not turned on. That is, the transmitter remains off, and no voice is transmitted by the wireless communication device.

  When the user of the wireless communication apparatus wants to transmit voice, particularly the real voice, the user speaks the real voice and inputs the real voice to the microphone. When the user speaks, the VOX circuit 100 performs VOX transmission and turns on the power of the transmitter. Accordingly, the user can transmit the real voice by transmitting a real voice by the transmitter.

However, there is a case where sound that is not intended by the user is input to the microphone 2. The sound not intended by the user is, for example, a sound (mechanical collision sound) generated when a wireless communication device having the VOX circuit 100 collides with something outside or when a mechanical switch provided in the wireless communication device is operated. is there. If such a sound not intended by the user is input to the microphone 2, the sound may cause VOX transmission, and the transmitter may be turned on.
As a method of transmitting when the condition that the threshold value is continuously exceeded for a certain period of time is satisfied so that it does not react to a single collision sound, if a mechanical collision sound occurs once, Since a reverberant sound is generated by the natural frequency, the threshold value may be detected continuously exceeding the predetermined time.
That is, the transmitter may be turned on regardless of the user's intention.

  Transmitter operations that do not rely on the user's intention waste power. In particular, in a portable wireless communication device driven by a battery, wasteful consumption of power leads to a reduction in available time and lowers the convenience of the wireless communication device.

In addition, the operation of the transmitter that is not intended by the user involves transmission of unintended voice (such as noise).
Reception of noise or the like may cause discomfort to the receiver.
In addition, the receiver may not be able to clearly determine whether the received noise is mere noise or has been accompanied by noise such as noise during transmission of the voice intended by the sender. . That is, unintentional transmitter actions involving unintended voice (such as noise) transmissions may confuse the recipient.

  In view of this, a method has been proposed for distinguishing between a voice intended by a user and a noise not intended by the user. As a method for distinguishing between voice due to the user's intention and noise not according to the user's intention, for example, whether or not the voice input based on the measurement result is a voice according to the user's intention is measured. Is known (for example, Patent Document 1).

JP 2006-38894 A

  However, the method described in Patent Document 1 requires a configuration for performing speech frequency measurement and speech discrimination based on frequency analysis results. The addition of the configuration causes a problem that costs and power consumption increase.

  For example, when a configuration for performing speech frequency measurement and speech discrimination based on the frequency analysis result is provided in a separate device, the number of parts increases and the cost of the wireless communication device increases, and the device becomes complicated. Increase in size. In addition, since the configuration for performing voice frequency measurement and voice discrimination based on the frequency analysis result consumes power, the power consumption of the VOX increases, and power saving performance is one of the advantages of VOX. Decreases.

In addition, there is a method of performing a configuration for performing speech frequency measurement and speech discrimination based on the frequency measurement result by software processing. In this case, frequency measurement processing and speech discrimination processing based on the measurement result As a result, the processing load on the CPU 5 increases, so a CPU having high processing capability is required.
In general, the cost of a CPU increases according to the processing capacity. For this reason, the cost of the VOX circuit 100 increases when the voice is discriminated by software processing. In addition, the CPU tends to increase the power consumption according to the processing capacity. For this reason, the power consumption of the VOX circuit 100 increases when the voice is determined by software processing.

  The above-described problems are not limited to the VOX circuit 100 in the wireless communication apparatus, but may be problems that may occur in all devices that perform operation control based on voice detection.

  An object of the present invention is to distinguish between a voice intended by a user and a noise not intended by the user without causing an increase in cost and power consumption.

  In order to solve the above-mentioned problem, an invention according to claim 1 is an operation control device for switching ON / OFF of a power supply of a device based on an output signal of a microphone, wherein the output signal is divided into predetermined time units. A value of the output signal corresponding to the volume of the sound is calculated for each predetermined time unit based on the output signal, the value of the output signal is compared with a predetermined threshold, and the value of the output signal is When the threshold value is exceeded, the value of the output signal is compared with the value of the output signal of a predetermined time unit immediately before the value, and the value of the output signal is continuously updated for the first predetermined number of times in the predetermined time unit. If the value of the output signal does not fall below the value of the immediately preceding predetermined time unit and the value of the output signal exceeds the threshold value continuously for the second predetermined number of times in the predetermined time unit, the power supply of the device is turned off. A control unit to turn on, Characterized in that it obtain.

  The invention according to claim 2 is the operation control apparatus according to claim 1, wherein the predetermined time unit is a predetermined value immediately before the value of the output signal continuously for the first predetermined number of times. It is a time unit that can distinguish between a real voice and a mechanical collision sound based on whether or not the value of the output signal in a time unit is below, and the first predetermined number of times is continuous more than the first predetermined number of times. The number of times that the real voice and the mechanical collision sound can be distinguished based on whether or not the value of the output signal is lower than the value of the output signal of a predetermined time unit immediately before the output signal value.

  A third aspect of the present invention is the operation control apparatus according to the first or second aspect, wherein the operation control apparatus is a VOX control apparatus.

  According to a fourth aspect of the present invention, there is provided an operation control method for switching ON / OFF of a power supply of a device based on an output signal of a microphone, wherein the output signal is divided by a predetermined time unit, and based on the divided output signal Calculating the value of the output signal corresponding to the volume of the sound, comparing the value of the output signal in the predetermined time unit with a predetermined threshold, and when the value of the output signal exceeds the threshold A step of comparing the value of the output signal with the value of the output signal in a predetermined time unit immediately before the output signal, and the value of the output signal is continuously changed to a predetermined value immediately before the first predetermined number of times in the predetermined time unit. A step of turning on the power of the device when the value of the output signal does not fall below the value of the output signal in time units and the value of the output signal exceeds the threshold value continuously for the second predetermined number of times in the predetermined time unit Special features To.

  According to the present invention, it is possible to distinguish between voice caused by the user's intention and noise not intended by the user without causing an increase in cost and power consumption.

It is a block diagram which shows the main structures of the VOX circuit which is one Embodiment of this invention. It is explanatory drawing which shows an example of the output signal produced by a real voice. It is a figure which shows an example of the output signal produced by the click sound of a mechanical switch. It is a figure which shows an example of the output signal produced by the collision sound produced when an apparatus collides with an external structure. It is explanatory drawing which shows an example which divided | segmented the output signal produced by the real voice of FIG. 2 in a unit of 4 [ms]. It is explanatory drawing which shows an example which divided | segmented the output signal of the rise P1 vicinity shown in FIG. 3 in 4 [ms] unit. It is a flowchart which shows an example of the flow of a VOX process by the VOX circuit of this embodiment. 1 is a block diagram showing a main configuration of a conventional VOX circuit 100. FIG.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing a main configuration of a VOX circuit 1 according to an embodiment of the present invention.
A VOX circuit 1 according to an embodiment of the present invention includes a microphone 2, an amplifier 3, a detection smoothing circuit 4, and a CPU 5, similarly to the VOX circuit 100 shown in FIG. 8. The microphone 2, amplifier 3, detection / smoothing circuit 4 and CPU 5 included in the VOX circuit 1 are the same as the microphone 2, amplifier 3, detection / smoothing circuit 4 and CPU 5 shown in FIG.

FIG. 2 is an explanatory diagram showing an example of an output signal generated by the real voice.
The microphone 2 outputs an output signal as shown in FIG.

The output signal is also output by a sound other than the real voice, for example, a mechanical collision sound. Hereinafter, an example of a mechanical collision sound is shown in FIGS.
FIG. 3 is a diagram illustrating an example of an output signal generated by a click sound of a mechanical switch.
In the waveform of the output signal shown in FIG. 3, the rising P1 is a sound when the mechanical switch is pressed, and the rising P2 is a sound when the mechanical switch is released from the pressed state and returned.
FIG. 4 is a diagram illustrating an example of an output signal generated by a collision sound generated when a device collides with an external structure.
These mechanical impact sounds have a characteristic that reverberation occurs due to the natural frequency of the casing, and the initial amplitude gradually attenuates.

  As shown in FIG. 2, FIG. 3 and FIG. 4, in addition to the output signal by the real voice, the output signal generated by the click sound accompanying the pressing operation of the mechanical switch or the collision sound generated when the device collides with the external structure. Even for an output signal that does not depend on the real voice, such as an output signal that occurs, the value of the output signal may exceed a predetermined threshold B.

Here, the change in the value of the output signal when the waveform of the output signal rises is divided by a predetermined time unit (for example, 4 [ms] unit), and the value of the divided output signal is compared with the value of the output signal immediately before To do.
FIG. 5 is an explanatory diagram showing an example in which the output signal generated by the real voice of FIG. 2 is divided in units of 4 [ms].
FIG. 6 is an explanatory diagram showing an example in which the output signal near the rising edge P1 shown in FIG. 3 is divided in units of 4 [ms].
As shown in FIG. 6, in the case of the rising edge P1 of the waveform of the output signal, the value of the output signal peaks at the rising edge and continues to attenuate thereafter. The pattern of the output signal that peaks at the time of rising and continues to attenuate thereafter is not limited to the rising P1, but is the output signal generated by the single collision sound that occurs when the device shown in FIG. 4 collides with the external structure. Is the same. That is, the value of the output signal when the output signal generated by the click sound accompanying the pressing operation of the mechanical switch or the output signal generated by the collision sound generated when the device collides with the external structure is divided by a predetermined time unit, After the rise of the output signal, the reverberation effect due to the natural frequency of the housing continues to be smaller than the value of the previous output signal.

  On the other hand, as shown in FIG. 5, in the case of an output signal based on the real voice, the value of the output signal is not always smaller than the value of the previous output signal. For example, as shown in FIG. 5, the values of the output signal of 8 [ms] and the output signal of 16 [ms] are larger than the value of the output signal immediately before that.

The VOX circuit 1 of the present embodiment performs processing for determining whether or not to perform VOX transmission based on the difference in the change pattern of the value of the output signal when the output signal is divided into predetermined time units.
First, the CPU 5 divides the output signal every predetermined time (4 [ms]). Then, the output signal divided in predetermined time units is A / D converted. The digitally converted output signal can be treated as a value, and the value of the output signal corresponds to the volume of sound input to the microphone 2.

Next, the CPU 5 determines whether or not the value of the output signal is lower than the value of the previous output signal continuously for a predetermined number of times (first predetermined number of times).
“The value of the output signal is not lower than the value of the previous output signal for a predetermined number of times (first predetermined number of times) continuously” means that the value of the output signal divided by a predetermined time unit is the previous output signal. This indicates that the value of the output signal in a predetermined time unit that is less than this value is not continuous for a predetermined number of times (first predetermined number of times).

  For example, a case where 3 [times] is used as the first predetermined number of times will be described. In the case of the output signal shown in FIG. 6, the value of the output signal is lower than the value of the previous output signal in the order of R2 for R1, R3 for R2, and R4 for R3. Is determined to have fallen below 3 consecutive times. On the other hand, in the case of the output signal shown in FIG. 5, since there are output signal values in which the value of the output signal exceeds the value of the previous output signal, such as Q2 for Q1 and Q4 for Q3, the CPU 5 It is determined that the value of the output signal is not lower than three consecutive times.

  Further, the CPU 5 determines whether or not the value of the output signal exceeds the threshold value continuously for a predetermined number of times (second predetermined number).

  “The output signal value exceeds the threshold value continuously for a predetermined number of times (second predetermined number of times)” means that the output signal value delimited in a predetermined time unit is output in a predetermined time unit exceeding the threshold value. It indicates that the value of the signal continues for a predetermined number of times (second predetermined number of times). For example, when 5 [times] is used as the second predetermined number of times and the output signal value divided in units of 4 [ms] is handled, the value of the output signal continuous over 20 [ms] Indicates that each exceeds the threshold.

  In the present embodiment, the predetermined time unit is 4 [ms], the first predetermined number of times is 3 [times], and the second predetermined number of times is 5 [times]. The user can set arbitrary values for the predetermined number of times and the second predetermined number of times. The values set in the predetermined time unit, the first predetermined number of times, and the second predetermined number of times are the output signal generated by the click sound accompanying the pressing operation of the mechanical switch or the collision that occurs when the device collides with an external structure. It is desirable that the output signal generated by the sound is set within a range that can be distinguished from the real voice.

  In other words, the predetermined time unit is a voice and a mechanical collision sound based on whether the value of the output signal is continuously lower than the value of the output signal of the predetermined time unit immediately before the first predetermined number of times. The first predetermined number of times is based on whether the value of the output signal is continuously lower than the value of the output signal of the predetermined time unit immediately before the first predetermined number of times. Therefore, it is desirable that the number of times can distinguish between a real voice and a mechanical collision sound.

  The user can adjust the volume (sensitivity) of the sound that triggers VOX transmission by changing the threshold setting. For example, as a high sensitivity setting, by setting a smaller threshold with respect to a predetermined threshold, the power can be turned on with a smaller sound than in the case of operation control based on the predetermined threshold. On the contrary, by setting the threshold value to be a larger threshold value than the predetermined threshold value as the low sensitivity setting, it is possible to require a louder sound than in the case of operation control based on the predetermined threshold value to turn on the power. .

The control of VOX transmission based on the value of the output signal divided in predetermined time units according to the present embodiment is particularly effective in the high sensitivity setting. This is because, in the case of the high sensitivity setting, compared to the case of other sensitivity settings, the output signal generated by the click sound generated when the mechanical switch is pressed and the collision sound generated when the device collides with an external structure. This is because the possibility that the value of the output signal generated by the above will exceed the threshold value increases. In the conventional method of performing VOX transmission only when the value of the output signal exceeds the threshold value, the above-described problem is more likely to occur in the case of high sensitivity setting. In this embodiment, even in the case of high sensitivity setting, VOX transmission is performed by an output signal generated by a click sound caused by a press operation of a mechanical switch or a collision sound generated when a device collides with an external structure. There is no.
The present embodiment is not limited to the high sensitivity setting, and VOX transmission can be controlled even when other sensitivity settings are used, and the effect is exhibited.

Hereinafter, an example of processing in the case of high sensitivity setting will be shown.
FIG. 7 is a flowchart showing an example of the flow of VOX processing by the VOX circuit 1 of the present embodiment.
First, the sensitivity is set to a high sensitivity setting by a user operation (step S1). Thereafter, when sound is input from the microphone 2, an output signal is input to the CPU 5 through amplification by the amplifier 3 and smoothing by the detection smoothing circuit 4, and the CPU 5 performs A / D conversion on the output signal (step S2). In step S2, the CPU 5 performs A / D conversion on the output signal in predetermined time units to obtain the value of the output signal.

  After the processing in step S2, the CPU 5 compares the digitized output signal value with a threshold value. Then, the CPU 5 determines whether or not the value of the digitized output signal exceeds a threshold value (step S3).

  In step S3, when the value of the output signal exceeds the threshold value (step S3: YES), the CPU 5 determines whether or not the value of the output signal is lower than the value of the previous output signal for three consecutive times (step S3). S4). In step S4, when the value of the output signal is not lower than the value of the previous output signal for three consecutive times (step S4: YES), the CPU 5 determines whether the value of the output signal exceeds the threshold value for five consecutive times. Determine. (Step S5). In step S5, when the value of the output signal exceeds the threshold value five times continuously (step S5: YES), the CPU 5 performs VOX transmission (step S6).

  When the value of the output signal does not exceed the threshold value at step S3 (step S3: NO), when the value of the output signal falls below the threshold value three times continuously at step S4 (step S4: NO), or five times at step S5 If the value of the output signal does not continuously exceed the threshold value (step S5: NO), the CPU 5 determines whether the input of the output signal continues after the output signal of the predetermined time unit for which the determination has been completed (step S5). S7). When the input of the output signal continues in step S7 (step S7: YES), the CPU 5 sets the next predetermined time unit output signal as the processing target (step S8), and returns to the processing of step S2. If the input of the output signal has not continued in step S7 (step S7: NO), the CPU 5 ends the process.

  The flowchart shown in FIG. 7 includes the procedure for setting the high sensitivity in step S1, but the high sensitivity setting is not necessarily required, and other sensitivity settings may be used.

As described above, according to the present embodiment, when the value of the output signal is not lower than the value of the previous output signal for three consecutive times and the value of the output signal exceeds the threshold value for five consecutive times, the CPU 5 VOX transmission is performed. Thus, the VOX circuit 1 can perform VOX transmission when the microphone 2 receives a real voice input. The VOX circuit 1 does not perform VOX transmission when the microphone 2 receives an input of a click sound of a mechanical switch or a collision sound generated when a device collides with an external structure. That is, the VOX circuit 1 according to the present embodiment is configured such that the voice intended by the user, such as the real voice, and the noise not caused by the user, such as a click sound of the mechanical switch or a collision sound generated when the device collides with an external structure. And the presence or absence of VOX transmission can be controlled.
In addition, the present embodiment is used to distinguish between voices such as the real voice, such as the voice of the user, and noises not caused by the user's intention, such as a click sound of a mechanical switch and a collision sound generated when a device collides with an external structure. In order to control the presence or absence of VOX transmission based on this, no additional hardware is required for the conventional configuration. Thus, it is possible to control the presence or absence of VOX transmission by distinguishing between voice intended by the user and noise not intended by the user without causing an increase in cost and power consumption.

  Furthermore, the predetermined time unit is a voice and a mechanical collision sound based on whether the value of the output signal is continuously lower than the value of the output signal of the predetermined time unit immediately before the first predetermined number of times. The first predetermined number of times is based on whether the value of the output signal is continuously lower than the value of the output signal of the predetermined time unit immediately before the first predetermined number of times. Therefore, the CPU 5 can distinguish well whether the output signal is a voice intended by the user or a noise not intended by the user. it can.

  Furthermore, since this embodiment is the VOX circuit 1 that controls VOX transmission, it is possible to satisfactorily suppress noise transmission in the wireless communication apparatus.

  The above description of the embodiment shows an example of the present invention, and does not limit the embodiment of the present invention. The configuration can be changed without departing from the characteristics of the present invention.

  For example, the present invention is not limited to the VOX circuit 1 of the wireless communication apparatus, but can be applied to other types of devices that perform operation control based on voice detection. As an example of another device to which the present invention is applied, there is a voice recorder or the like that performs activation / termination control based on the distinction between voice and noise according to the present invention.

2 Microphone 3 Amplifier 4 Detection and smoothing circuit 5 CPU

Claims (4)

An operation control device for switching on / off the power of a device based on an output signal of a microphone,
The output signal is divided by a predetermined time unit, the value of the output signal corresponding to the loudness is calculated for each predetermined time unit based on the divided output signal, and the value of the output signal is predetermined. Compared with a threshold value, when the value of the output signal exceeds the threshold value, the value of the output signal is compared with the value of the output signal of a predetermined time unit immediately before the first predetermined value in the predetermined time unit. The value of the output signal does not fall below the value of the output signal in a predetermined time unit immediately before the number of times, and the value of the output signal continues continuously for the second predetermined number of times in the predetermined time unit. An operation control apparatus comprising: a control unit that turns on the power of the device when the threshold value is exceeded. The predetermined time unit is determined based on whether the value of the output signal is continuously lower than the value of the output signal of the predetermined time unit immediately before the first predetermined number of times. Is a unit of time that can be distinguished
The first predetermined number of times is determined based on whether the value of the output signal is continuously lower than the value of the output signal of the predetermined time unit immediately before the first predetermined number of times. The operation control device according to claim 1, wherein the operation control device is a number of times that can be distinguished.   The operation control apparatus according to claim 1, wherein the operation control apparatus is a VOX control apparatus. An operation control method for switching ON / OFF of the power supply of a device based on a microphone output signal,
Dividing the output signal by a predetermined time unit, and calculating a value of the output signal corresponding to the loudness based on the divided output signal;
Comparing the value of the output signal in the predetermined time unit with a predetermined threshold;
A step of comparing the value of the output signal with the value of the output signal in a predetermined time unit immediately before the value when the value of the output signal exceeds the threshold;
The value of the output signal does not fall below the value of the output signal of the predetermined time unit immediately before the first predetermined number of times in the predetermined time unit, and the second predetermined number of times in the predetermined time unit And a step of turning on the power of the device when the value of the output signal continuously exceeds the threshold value.

Images