JPH10207470A - Active noise reduction device and method therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an active noise reduction device which can reduce only noise. SOLUTION: A control device receives a microphone signal M and a rotation sensor signal R, and separates noise component by subtracting a component based on audio signal. A from the microphone signal M in a separation part 14, and outputs the noise component as a separated signal P. And, a phase adjustment is performed between the rotation sensor signal R and the separated signal P by a driving signal calculation part 15 which updates an adaptive filter for minimizing the separated signal P, filtering it with the updated filter characteristic, making a driving signal Y, making a summed signal T (=A+Y) by adding the audio signal to the driving signal Y, and outputting the summed signal T to a loudspeaker 1.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an active noise reduction device and a noise reduction method for reducing noise by causing control noise and noise to interfere with each other.

[0002]

2. Description of the Related Art Conventionally, in this type of active noise reduction apparatus, in order to achieve effective noise reduction, a control sound having the same frequency, the same amplitude and the opposite phase to the noise is generated by a sound generating means such as a speaker. Must be caused by Therefore, Filtered-X-LMS (Least Me
When a noise generation source is, for example, an automobile engine, a control method called “square control” is applied to an active noise reduction device, and a rotation sensor signal (reference signal) of the noise generation source and a microphone signal obtained from a noise detection microphone are used. An optimum drive signal for the speaker is calculated based on the speaker.

[0003] Such active noise reduction devices include:
For example, there is one disclosed in Japanese Patent Application Laid-Open No. Hei 5-11777, in which the transfer characteristics of a transfer system such as a sound field existing between a speaker and a microphone are estimated in advance using a music source. Then, using this transfer characteristic, the adaptive digital filter is changed so that the canceling sound detected by the microphone is minimized.

[0004]

However, in the active noise reduction device disclosed in Japanese Patent Laid-Open No. Hei 5-11777,
Although the sound detected by the microphone includes useful sounds such as music output from audio devices, all of them are treated as noise, including the sound components that humans want to hear. There was a disadvantage that the sound was muted.

The present invention has been made in view of the above problems, and an object of the present invention is to provide an active noise reduction device and a noise reduction method capable of reducing only noise.

[0006]

In order to solve the above-mentioned problems, an active noise reduction device according to the first aspect of the present invention comprises a sound detection means for detecting a sound generated indoors as a sound detection signal; A separating unit that removes a component based on the audio signal from the sound detection signal detected by the sound detecting unit to obtain a noise component to be silenced; a sound generating unit that generates a control sound for silencing the noise component; And a drive signal generating means for generating a drive signal for driving the sound generating means in accordance with the noise component.

In such an active noise reduction device, a sound detecting means such as a microphone or a vibration sensor detects a sound generated in a room such as an automobile, an aircraft, a ship, or a railroad vehicle as a sound detection signal. The separating means removes a component based on the audio signal from the sound detection signal to obtain a noise component to be silenced. Here, the audio signal refers to a signal output from an audio device such as a television, a radio, or a video, and refers to a signal related to music, announcements, narration, and the like. Further, the component based on the audio signal may be the audio signal itself output from the audio device, or may be a signal obtained by applying some processing to the audio signal. The sound generating means is driven by the drive signal generated by the drive signal generating means, and generates a control sound for silencing a noise component.

According to such an active noise reduction device, since only the noise component after removing the component based on the audio signal is to be silenced, only the noise can be reduced, and the music or the like that one wants to hear is protected. Is obtained. According to a second aspect of the present invention, the active noise reduction device according to the first aspect further includes a transfer characteristic estimating unit that estimates a transfer characteristic between the sound detection signal and the drive signal. The separating means obtains a noise component to be silenced by subtracting a signal obtained by multiplying the audio signal by a transfer characteristic from the sound detection signal, and the sound generating means further includes the audio as the audio sound in addition to the control sound. It is configured to generate a sound corresponding to a signal.

Such an active noise reduction device considers the case where the effect of the transfer characteristic between the sound detection signal and the drive signal (that is, the transfer characteristic between the sound detection means and the sound generation means) is large. . That is, when the audio sound (sound corresponding to the audio signal) generated from the sound generating means reaches the sound detecting means, it is detected as a signal multiplied by the transfer characteristic. Therefore, the separating means obtains the noise component by subtracting a signal obtained by multiplying the audio signal by the transfer characteristic from the sound detection signal.

According to such an active noise reduction device, especially when the influence of the transfer characteristic is large, the noise component is extracted from the sound detection signal in consideration of the transfer characteristic. The effect that music and the like that a person wants to listen to is surely protected.

According to a third aspect of the present invention, there is provided the active noise reduction apparatus according to the first aspect, further comprising an inverse characteristic of a transfer characteristic between the sound detection signal and the drive signal (hereinafter, referred to as an “inverse transfer characteristic”). The separation means obtains a noise component to be silenced by subtracting the audio signal from the sound detection signal, and the sound generation means includes the control sound. In addition, a sound corresponding to a signal obtained by multiplying the audio signal by a reverse transfer characteristic is generated as an audio sound.

Such an active noise reduction device also takes into consideration the case where the above-mentioned transmission characteristics have a large effect. That is,
If the sound according to the audio signal is directly generated by the sound generating means, when the sound is detected by the sound detecting means, that is, when a human hears the sound, the sound is detected or multiplied by the transmission characteristic. Will be heard. In view of this point, in the present invention, a sound corresponding to a signal obtained by multiplying an audio signal by a reverse transfer characteristic is generated as an audio sound,
When the sound is detected by the sound detecting means, that is, when the sound is heard by the human ear, the sound itself is multiplied by the transfer characteristic, so that the sound itself corresponding to the audio signal can be detected or heard.

In this case, since the sound detection signal includes a sound corresponding to the audio signal, the separating means obtains a noise component by subtracting the audio signal from the sound detection signal. According to such an active noise reduction device, particularly when the influence of the transfer characteristics is large, the audio sound is generated in consideration of the transfer characteristics, so that the listener can hear good audio sounds. The effect is obtained.

According to a fourth aspect of the present invention, there is provided the active noise reduction apparatus according to any one of the first to third aspects, further comprising a reference signal generating means for generating a reference signal that follows a period of the noise. The drive signal generating means includes an adaptive filter capable of variably setting a phase and a gain, and variably sets a filter characteristic of the adaptive filter so as to minimize the noise component, and generates the drive signal from the reference signal. Features.

In such an active noise reduction device, since the period of the drive signal follows the period of the noise source, the noise can be suitably reduced even in devices, automobiles, trains, and the like in which the period of the noise source varies greatly. Is obtained. In the active noise reduction apparatus according to any one of claims 1 to 4, for example, the CPU of the computer reads an active noise reduction control program recorded on a recording medium such as a ROM and executes the processing. May be configured. At this time, the control program causes the computer to detect a sound generated in the room, remove a component based on an audio signal that is not a sound-reduction target from the detection result to obtain a noise component that is a sound-reduction target, and drives the sound generation unit. It is described that a drive signal is generated in accordance with the noise component, and the sound signal is driven by the drive signal to generate a control sound for silencing the noise component.

According to a fifth aspect of the present invention, there is provided a noise reduction method comprising: detecting a sound generated in a room as a sound detection signal; removing a component based on an audio signal from the sound detection signal to obtain a noise component to be silenced; A drive signal for driving the means is created in accordance with the noise component, and the control signal for silencing the noise component is generated by driving the sound generating means with the drive signal.

According to such a noise reduction method, only the noise component after removing the component based on the audio signal is set as a noise reduction target, so that only the noise can be reduced.
The effect of protecting music and the like that one wants to hear is obtained. The noise reduction method according to claim 6, wherein a sound generated indoors is detected as a sound detection signal, a transfer characteristic between the sound detection signal and the drive signal is estimated, and the sound detection signal is converted to the audio signal. A noise component to be silenced is obtained by subtracting the signal obtained by multiplying the transfer characteristic, a drive signal for driving a sound generating unit is created according to the noise component, and the drive signal and the audio signal are created according to the drive signal and the audio signal. By driving the sound generating means, a control sound and an audio sound for silencing the noise component are generated.

According to this noise reduction method, especially when the influence of the transfer characteristic is large, the noise component is extracted from the sound detection signal in consideration of the transfer characteristic, so that only the noise can be accurately reduced. Thus, the effect that music and the like that a person wants to listen to can be surely protected.

According to a seventh aspect of the present invention, a noise reduction method detects a sound generated in a room as a sound detection signal, estimates a reverse transfer characteristic between the sound detection signal and the drive signal, and calculates a reverse transfer characteristic from the sound detection signal. A noise component to be silenced is obtained by subtracting the audio signal, a drive signal for driving a sound generating unit is created according to the noise component, and the drive signal and the audio signal are multiplied by the inverse transfer characteristic. The sound is generated from the sound generating means in accordance with the signal.

According to such a noise reduction method, particularly when the influence of the transfer characteristic is large, the audio sound is generated in consideration of the transfer characteristic, so that the listener can hear a good audio sound. The effect is obtained.

[0021]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below with reference to the drawings. [First Embodiment] FIG. 1 is a system configuration diagram of an active noise reduction apparatus of the present invention mounted on an automobile.

The active noise reduction device mainly includes a speaker 1, a rotation sensor 2, a microphone 3, and a control device 10. The speaker 1 is a sound generating means of the present invention, and generates a control sound in a space in a vehicle compartment to which noise is transmitted from a noise generating source such as an engine (abbreviated as E / G). In addition, as the sound generating means, in addition to the speaker, for example, a vibrating actuator (not shown) may be attached to a vibrating portion such as the roof panel 4, the dash panel 5, or the floor panel 6 existing in the path where noise and vibration are transmitted from the noise source. ) May be used to reduce noise by vibration of the roof panel 4, the dash panel 5, or the floor panel 6 by the vibration actuator.

The rotation sensor 2 is a reference signal detecting means of the present invention, outputs a rotation sensor signal R (reference signal) as a reference for driving the speaker 1, and has a noise source E / In synchronization with the explosion of G, a predetermined number of pulses are output per E / G rotation. As the reference signal detecting means, for example, a vibration sensor may be used in addition to the rotation sensor. When using a vibration sensor,
It may be mounted on a position where the state of the noise source can be detected and the sound emitted from the speaker 1 or the generated vibration is not intervened, for example, an E / G body, an E / G mount, or the like.

The microphone 3 is a sound detecting means of the present invention, which detects a state of a sound generated in the indoor space and outputs a microphone signal M (sound detection signal). The microphone 3 may be mounted at a position where the sound should be muted, such as at the occupant's ear. As the sound detecting means, for example, a vibration sensor may be used other than the microphone. When a vibration sensor is used, the vibration sensor may be mounted on a vibrating part having a correlation with the sound at the position to be silenced, for example, a roof panel 4 or a floor panel 6 in a vehicle.

The control device 10 generates a drive signal Y for driving the speaker 1 and controls the drive signal Y. The input processing unit 11, the input processing unit 12, the output processing unit 13, and the separation unit 14, drive signal calculator 15, adder 1
6, a changeover switch 17, and a transfer characteristic C estimation processing unit (hereinafter, also referred to as "C estimation processing unit") 18. As the control device 10, for example, a computer such as a microprocessor or a digital processor can be used, and a processing procedure (described later) stored in a built-in ROM is executed by a CPU built in the microprocessor.
Each operation of the separation unit 14, the drive signal calculation unit 15, the addition unit 16, the changeover switch 17, and the C estimation processing unit 18 is controlled.

The input processing unit 11 is constituted by a known circuit that removes noise in the microphone signal M by a low-pass filter and amplifies the microphone signal M at a predetermined amplification factor. The microphone signal M that has passed through the input processing unit 11 is input to the separation unit 14. The input processing unit 12 includes a known circuit that removes noise of the rotation sensor signal R from the rotation sensor 2 with a low-pass filter and amplifies the rotation sensor signal R at a predetermined amplification factor. The rotation sensor signal R that has passed through the input processing unit 12 is
5 is input.

The output processing unit 13 removes noise of the addition signal T output from the addition unit 16 by a low-pass filter,
In order to drive the speaker 1, the addition signal T at a predetermined amplification rate
Is configured by a well-known circuit that amplifies the current. The addition signal T that has passed through the output processing unit 13 is input to the speaker 1.

The separation unit 14 generates a component based on the audio signal A, in this case, a signal (A) obtained by multiplying the audio signal A by a transfer characteristic C between the speaker 1 and the microphone 3.
C) and the microphone signal M input from the input processing unit 11
Is input to the adder 14a, and the adder 14a subtracts the former from the latter to calculate a noise component to be silenced, and uses the result as a separated signal P (= M−A · C). 15 is output. Note that the audio signal A is a signal output from the audio device 7.

The drive signal calculation section 15 calculates the phase and amplitude of the drive signal of the speaker 1 and creates and outputs a drive signal Y having the phase and amplitude based on the rotation sensor signal R. The adder 16 adds the drive signal Y input from the drive signal calculator 15 and the audio signal A input from the audio device 7 and outputs the result to the output processor 13 as an addition signal T (= A + Y).

Here, the details of the drive signal calculator 15 will be described with reference to FIG. FIG. 2 is a block diagram illustrating the function of the drive signal calculation unit. As shown in FIG. 2, the drive signal calculation section 15 includes an adaptive filter section 15a, a pre-filter section 15b, and an adaptive filter update processing section 15c.

The adaptive filter section 15a has a filter characteristic H whose phase and gain can be varied, and generates a drive signal Y based on a rotation sensor signal R output from the rotation sensor 2. The pre-filter unit 15b corrects a delay of the separated signal P of the microphone 3 with respect to the drive signal Y. This delay is caused by the transfer characteristic C between the separation signal P and the drive signal Y. For this reason,
The filter processing of the filter characteristic C0 equal to the transfer characteristic C is performed by the pre-filter unit 15b.

The adaptive filter update processing section 15c variably sets the filter characteristic H of the adaptive filter section 15a, more specifically, the phase / gain so that the separated signal P is minimized. This adaptive filter update processing unit 15c
Sets the filter characteristic H of the adaptive filter unit 15a to Fil
The value is obtained by a theoretical formula described later used in a control method called “tered-X-LMS control”, and the value is repeatedly updated. This update processing is executed under control without divergence in which the transfer characteristic C is fixed. Thus, the separated signal P, which is a noise component, is minimized.

Here, the output of the pre-filter unit 15b is X
Assuming that d and the update interval (constant) are μ1, the value after update of the adaptive filter unit 15a is expressed by Equation 1 using the values of these parameters and the value of the filter characteristic H before the update of the adaptive filter unit 15a. Can be expressed as

[0034]

(Equation 1)

Next, a method for measuring the transfer characteristic C between the separation signal P and the drive signal Y will be described. Drive signal calculator 15
Each microphone signal when the changeover switch 17 provided between the switch and the adder 16 is on, that is, when the drive signal output is on, and when this changeover switch 17 is off, that is, when the drive signal output is off, M1, M2, and the separation signal is P
Assuming that P1 and P2, the C estimation processing unit 18 that estimates the transfer characteristic C can estimate the transfer characteristic C from the separated signals P1 and P2 and the drive signal Y.

The separation signals P1 and P2 represent the engine noise
Then

[0037]

(Equation 2)

[0038]

(Equation 3)

Is represented by Here, assuming Equation 2−Equation 3, the transfer characteristic C becomes

[0040]

(Equation 4)

It can be estimated from the difference (P1−P2) between the separated signals when the drive signal output is on and off, and the drive signal Y. Note that the transfer characteristic C changes when some disturbance is involved, and there is no auditory problem even if the drive signal Y is turned off. Further, in order to obtain Equation 4, the transfer functions W1 (when the drive signal output is on) and W2 (when the drive signal output is off) when the input is the drive signal Y and the output is the separation signals P1 and P2.

[0042]

(Equation 5)

[0043]

(Equation 6)

Then, a difference between the transfer function W1 and the transfer function W2 is obtained. Next, the control system of FIG. 3 is configured to calculate the transfer functions W1 and W2. The control system shown in FIG. 3 shows a portion related to the adaptive filter unit 15a of the control system shown in FIG. 2 in detail. Transfer function Wi of adaptive filter section 15a
An arbitrary initial value is given to (i = 1, 2), and the following error E is obtained.

[0045]

(Equation 7)

[0046]

(Equation 8)

In order to set the error E to 0, the following Filt is used.
The filter characteristic updating means 51 repeatedly updates the transfer function Wi according to the ered-X-LMS control theoretical formula.

[0048]

(Equation 9)

Finally, if the error is E0, the transfer function W1,
The transfer characteristic C of Expression 4 is obtained from W2, and the transfer characteristic of the pre-filter is updated. Next, specific processing contents of the control device 10 will be described based on the flowchart of FIG. FIG.
The control procedure shown in (1) is previously recorded in the form of a program language readable by a computer (in this case, a microprocessor) in a ROM serving as a recording medium, and is repeatedly executed at a constant cycle.

The microprocessor inputs the rotation sensor signal R and the microphone signal M in the input processing of step (hereinafter abbreviated as “S”) 1, and subsequently separates the noise component from the microphone signal M in the separation processing of S2. Separated signal P
Output as That is, a signal obtained by multiplying the audio signal input from the audio device 7 by the transfer characteristic C is subtracted from the microphone signal M to calculate and output a separated signal P (= M−C · A) as a noise component signal. .

Subsequently, the microprocessor adjusts the phase of the rotation sensor signal R and the separation signal P in the pre-filtering process in S3, and adapts so that the separation signal P is minimized in the filter characteristic updating process in S4. The filter characteristic H in the filter processing is updated.

After this processing, the microprocessor proceeds to S5
, The updated filter characteristic H
To generate a drive signal Y, and in the addition processing in S6, add the audio signal A to the drive signal Y to create an addition signal T (= A + Y).

Then, the addition signal T is output to the speaker 1 by the addition signal output processing in S7. Then, the microphone 3 inputs a signal in which the addition signal T is affected by the transmission system, that is, C · T (= C · A + C · Y), and outputs this as the microphone signal M. Thereafter, the microprocessor estimates the transfer characteristic C by performing the above-described procedure for estimating the transfer characteristic C in the transfer characteristic C estimation process in S8.

As described above, according to the present embodiment, since only the noise component after removing the component based on the audio signal is to be silenced, only the noise can be reduced, and music or the like that one wants to listen to can be reduced. Has the effect of being protected. In addition, especially when the influence of the transfer characteristics is large, the noise component is extracted from the microphone signal in consideration of the transfer characteristics. The effect of being reliably protected is obtained.

[Second Embodiment] FIG. 5 is a system configuration diagram of the active noise reduction device of the present invention mounted on an automobile.
The active noise reduction device according to the second embodiment differs from the first embodiment in the configuration and processing of the separation unit 24 and the addition unit 26, and also performs reverse transmission instead of the C estimation processing unit 18 in the first embodiment. Characteristic C ^-1 estimation processing unit (hereinafter also referred to as "C ^-1 estimation processing unit")
The configuration is substantially the same as that of the first embodiment except that the configuration is different from that of the first embodiment. Therefore, the same components are denoted by the same reference numerals, and description thereof will be omitted.

The adder 26 of the present embodiment includes a drive signal Y input from the drive signal calculator 15 and the inverse transfer characteristic processor 2
In 9, an audio signal A input from the audio device 7 is added to a signal obtained by multiplying the inverse transfer characteristic C ⁻¹ , and this is output to the output processing unit 13 as an added signal T (= Y + A · C ⁻¹ ). I do.

In this embodiment, a signal obtained by multiplying the audio signal A by the inverse transfer characteristic C ^-1 in addition to the drive signal Y is output from the speaker 1. When these signals reach the microphone 3 from the speaker 1, , And the transfer characteristic C are multiplied. That is, the signal obtained by multiplying the audio signal A by the inverse transfer characteristic C ^-1 is further multiplied by the transfer characteristic C, so that the original audio signal A
It is input to the microphone 3 as it is. For this reason, there is an effect that the listener can listen to the audio signal A in a state that is not affected by the transmission system such as the sound field. This effect is particularly useful when the effect of transfer characteristics such as a sound field is large.

The separating section 24 adds the audio signal A itself as a component based on the audio signal A and the microphone signal M input from the input processing section 11 to the adder 24a.
And the adder 24a subtracts the former from the latter to calculate the noise component to be silenced,
This is used as the separation signal P (= M−A) as the drive signal calculation unit 1
5 is output.

In the present embodiment, since the microphone 3 is mounted at substantially the same position as the listener's ear, the microphone signal M
Are substantially the same as those heard by the listener (that is, A). For this reason,
The component to be subtracted in the separation unit 24 may be the audio signal A itself.

The C ^-1 estimating section 28 obtains the transfer characteristic C in the same manner as the C estimating section 18 of the first embodiment, and then obtains the inverse transfer characteristic C ^-1 which is the reverse characteristic. It is. Therefore, a detailed description is omitted. The procedure of the specific processing contents of the control device 10 is the same as that of the flowchart of FIG.
This control procedure is recorded in advance in a ROM, which is a recording medium, in the form of a program language readable by a computer (in this case, a microprocessor), and is repeatedly executed at a constant cycle.

The microprocessor constituting the control device 10 inputs the rotation sensor signal R and the microphone signal M in the input processing of S1, as in the first embodiment. In the subsequent separation process of S2, a noise component is separated from the microphone signal M and output as a separated signal P. That is, the audio signal A itself input from the audio device 7 is
To calculate and output a separated signal P (= MA) as a noise component signal.

Subsequently, the microprocessor performs the same processing as in the first embodiment for the pre-filtering processing in S3, the filter characteristic updating processing in S4, and the adaptive filtering processing in S5. In the subsequent addition processing in S6, a signal obtained by multiplying the audio signal A by the inverse transfer characteristic C ⁻¹ is added to the drive signal Y to create an addition signal T (= A · C ⁻¹ + Y).

Then, the addition signal T (= A · C ⁻¹ + Y) is output to the speaker 1 by the addition signal output processing of S 7.
Thereafter, in the transfer characteristic C estimating process in S8, the microprocessor estimates the transfer characteristic C by performing the above-described procedure for estimating the transfer characteristic C, and further obtains an inverse transfer characteristic C- ¹ , which is an inverse characteristic thereof.

As described above, according to the present embodiment, since only the noise component after removing the component based on the audio signal is to be silenced, only the noise can be reduced, and the music or the like that one wants to hear can be reduced. Has the effect of being protected. In addition, particularly when the influence of the transfer characteristic is large, the audio sound is generated in consideration of the transfer characteristic, so that the effect that the listener can hear a good audio sound can be obtained.

It should be noted that the present invention is not limited to the above-described embodiment at all, and it is needless to say that various forms can be adopted as long as they belong to the technical scope of the present invention. For example, in the first embodiment, if the influence of the transfer characteristic is not particularly large, the separating unit 14 processes the audio signal A so as to subtract it from the microphone signal M without processing it. The configuration may be such that the signal itself and the drive signal Y are added to create an addition signal T and output this to the speaker 1.

In the first embodiment, the sound generating means is configured to output the control sound and the audio sound by using only one speaker 1. However, the sound generating means is replaced by the control sound generating speaker and the audio sound generating speaker. They may be separately output separately. In this case, the transfer characteristics between the control sound generation speaker and the microphone and the transfer characteristics between the audio sound generation speaker and the microphone are estimated, and the drive signal calculation unit 15 generates the drive signal. The former transfer characteristic may be used, and the latter transfer characteristic may be used when the separation unit 14 extracts a noise component from the microphone signal M. Note that a similar configuration may be adopted for the second embodiment.

[Brief description of the drawings]

FIG. 1 is a system configuration diagram of an active noise reduction device according to a first embodiment.

FIG. 2 is a block diagram illustrating functions of a drive signal calculation unit.

FIG. 3 is a block diagram partially showing the control system of FIG. 2;

FIG. 4 is a flowchart illustrating a processing procedure of a control device.

FIG. 5 is a system configuration diagram of an active noise reduction device according to a second embodiment.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Speaker, 2 ... Rotation sensor, 3 ... Microphone, 7 ... Audio device, 10 ... Control device, 14 ... Separation part, 15 ... Drive signal calculation part,
15a ... Adaptive filter unit, 15b ... Pre-filter unit, 15c ... Adaptive filter update processing unit, 16 ...
An addition unit, 17 a changeover switch, 18 a transfer characteristic C estimation processing unit;

Claims (7)

[Claims] 1. A sound detection means for detecting a sound generated in a room as a sound detection signal; and a noise component to be silenced by removing a component based on an audio signal from the sound detection signal detected by the sound detection means. Separating means, sound generating means for generating a control sound for silencing the noise component, and drive signal generating means for generating a drive signal for driving the sound generating means in accordance with the noise component. Active noise reduction device. 2. The active noise reduction device according to claim 1, further comprising: transfer characteristic estimating means for estimating a transfer characteristic between the sound detection signal and the drive signal; A noise component to be silenced is obtained by subtracting a signal obtained by multiplying the audio signal by a transfer characteristic from a detection signal, and the sound generation unit generates a sound corresponding to the audio signal as an audio sound in addition to the control sound. An active noise reduction device characterized by being generated. 3. The active noise reduction device according to claim 1, wherein an inverse transmission characteristic for estimating an inverse characteristic of a transmission characteristic between the sound detection signal and the drive signal (hereinafter, referred to as an “inverse transmission characteristic”). A characteristic estimating unit, wherein the separating unit obtains a noise component to be silenced by subtracting the audio signal from the sound detection signal, and the sound generating unit includes, in addition to the control sound, the audio signal as an audio sound. An active noise reduction device characterized by generating a sound corresponding to a signal obtained by multiplying a signal by a reverse transmission characteristic. 4. The active noise reduction device according to claim 1, further comprising: a reference signal generation unit configured to generate a reference signal that follows a period of the noise, wherein the drive signal generation unit includes a phase control unit. And an adaptive filter capable of variably setting a gain, wherein the drive signal is created from the reference signal by variably setting a filter characteristic of the adaptive filter so that the noise component is minimized. Reduction device. 5. A sound generated in a room is detected as a sound detection signal, a component based on an audio signal is removed from the sound detection signal to obtain a noise component to be silenced, and a driving signal for driving a sound generating means is generated. A noise reduction method, wherein the control signal is generated in accordance with a noise component, and a control sound for silencing the noise component is generated by driving the sound generation unit with the drive signal. 6. A sound generated in a room is detected as a sound detection signal, a transfer characteristic between the sound detection signal and the drive signal is estimated, and the audio signal is multiplied by the transfer characteristic from the sound detection signal. A noise component to be silenced is obtained by subtracting the combined signal, a drive signal for driving the sound generation unit is created according to the noise component, and the sound generation unit is driven according to the drive signal and the audio signal. Generating a control sound and an audio sound for silencing the noise component. 7. A sound generated in a room is detected as a sound detection signal, a reverse transfer characteristic between the sound detection signal and the drive signal is estimated, and the audio signal is subtracted from the sound detection signal to mute the sound. The target noise component is obtained, a drive signal for driving the sound generation unit is created according to the noise component, and the drive signal and the audio signal are multiplied by the inverse transfer characteristic to generate a drive signal from the sound generation unit. A noise reduction method characterized by generating sound.