JPH07240989A - Noise reduction headphone device - Google Patents

Abstract

PURPOSE:To prevent a noise reduction effect from being degraded due to the malfunction of coefficient correction by adaptive algorithm and to prevent the danger of generating new noise from the driver of a headphone by turning off or attenuating the output of an adaptive filter. CONSTITUTION:The adaptive filter 16 filtering-processes input from a first microphone 14, sends it to an adder 4 and adaptively controls filter characteristics at the time so as to minimize the input from a second microphone 15. To an addition circuit 4, filter output from the adaptive filter 16 and audio signals and, music signals, etc., through a signal input terminal 7 are supplied. An attachment/detachment detection part 61 detects whether or not a headphone case body 1 is attached to a head part in close contact. When the headphone case body 1 is not correctly attached to the head part of user, a muting circuit or attenuation circuit 63 is ON-controlled by an output control part 63 and addition output is not supplied to a speaker 5 as it is.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a noise reduction headphone device suitable for use in a place where external noise is severe.

[0002]

2. Description of the Related Art Generally, when a headphone device for listening to an audio signal such as voice or music is used in a place where there is a lot of external noise, the noise mixed in interferes with the listening of the voice or music. There is a problem that The following techniques are known to solve this problem.

The first technique is to change the frequency characteristic of a voice or music signal to make noise inconspicuous.
This technique has a drawback that it not only impairs the original voice and music, but also works only on specific noise.

As shown in FIGS. 5 and 6, the second technique is that external noise is picked up by a microphone 2 mounted outside the headphone housing 1, and the picked-up signal is sent to the equalizer 3. By adjusting the equalizer 3 to a signal whose amplitude is equal and opposite in phase to the noise mixed in the headphone 1, the equalizer 3 sends it to the speaker 5 of the headphone 1 via the adder 4 for reproduction. The noise reaching the human ear (ear canal) 6 is canceled. The adder 4 is supplied from the signal input terminal 7 with a voice signal, a music signal or the like to be heard.

FIG. 6 shows the configuration of FIG. 5 in the form of a block using a transfer function. External noise (noise) N from the input terminal 11 is acoustic-electrically converted by the microphone 2 having a transfer function M. Then, the transfer function is electrically supplied to the speaker 5 of the headphone 1 having a transfer function of H via the equalizer 3 having a transfer function of −γ.
The sound is converted and sent to the adder 12. This adder 12
The external noise (noise) N is supplied through the headphone housing 1 having the sound insulation characteristic F to be added and mixed in the internal space of the headphone, and the human ear (the ear canal 6) via the output terminal 13.
To reach.

In such a configuration, the addition output from the output terminal 13 is NF + NM (-γ) H = N (F-MγH), so the transfer function of the equalizer 3 is If −γ is adjusted so that γ = F / (H · M), then F−M · γ · H = 0,
Can cancel noise. Although not shown in FIG. 6, an electric signal of voice or music to be listened to is supplied between the equalizer 3 and the speaker 5 of the headphone 1 and superimposed. According to this, there is an advantage that it does not affect the original voice or music and, in principle, does not depend on the nature of noise.

However, in reality, if the equalizer 3 is not adjusted accurately, the effect of canceling noise is small, and even after the adjustment, the characteristics of the speaker 5 of the headphone 1 depend on the shape around the ear canal and the manner of wearing the headphones. (Mainly the electric-acoustic conversion characteristic) H may change and the effect may be reduced.

Next, as a third technique, as shown in FIGS. 7 and 8, the equalizer 3 having the configuration shown in FIGS. 5 and 6 is used.
It is known to replace the with an adaptive filter 16.
That is, in FIG. 7, a first microphone 14 for noise collection, which is provided outside the headphone housing 1 and located near the ear when the headphone is mounted, and inside the headphone housing 1 when the headphone is mounted. Headphones 1
Using the second microphone 15 provided between the speaker 5 and the ear canal, the adaptive filter 16 filters the input from the first microphone 14 and sends it to the adder 4, and at this time. Filter characteristics (transfer function −
γ) is adaptively controlled so that the input from the second microphone 15 is minimized. FIG. 8 is a block diagram represented by using the transfer function of each unit, and the first microphone 1
The transfer function of No. 4 is M ₁ , and the transfer function of the second microphone 15 is M ₂ . Note that other configurations are the same as those in FIGS.
Since it is the same as the above, the same reference numerals are given to the corresponding parts and the description thereof will be omitted. Further, a signal such as a voice or music to be listened to has no correlation with noise to be reduced or canceled, and does not affect the noise reduction operation, and therefore description thereof will be omitted.

According to this configuration, the signal from the adder 4 is sent to the speaker 5 of the headphone 1 to reduce the noise component, and the noise-reduced internal sound of the headphone is the second.
The microphone 15 collects the sound and sends it to the adaptive filter 16 as a so-called residual. The adaptive filter 16 learns to minimize the power of the residual signal and adjusts its own filter characteristic.

The internal structure of the adaptive filter 16 is shown in FIG.
As shown in FIG.
It consists of 2. The input from the first microphone 14 is supplied to the terminal 17, and as a so-called reference input x, via the input terminal 16a of the adaptive filter 16,
It is sent to the filter unit 21 and the adaptive algorithm unit 22. The output y from the filter unit 21 is taken out via the terminal 16b as the output of the adaptive filter 16 and sent to the adder 18. The adder 18 represents the acoustic mixing inside the headphones, and the external noise (noise)
Is transmitted to the adder 18 via the terminal 19 via the terminal 19 via the sound insulation characteristic functional block such as the headphone housing, and the adaptive filter output y is subtracted from the noise component d. A so-called noise residual e (= d−y) is obtained. The residual e from the adder 18 (collected by the second microphone 15) is supplied to the adaptive algorithm unit 22 via the terminal 16c. The adaptive algorithm unit 22 includes a filter unit 21.
Output y obtained by filtering the input x by changing the filter coefficient of
2), adaptive control is performed so as to minimize the power of the residual e.

According to the third technique, since the adjustment is performed by learning, the adjustment is accurate, the noise reduction effect can be enhanced, and the readjustment is performed even if the characteristics of the headphones change later. It has the advantage of being easy. Therefore,
Of the three techniques described above, the third technique is considered to be the most effective.

However, in the noise reduction using the adaptive filter described as the third technique, there is a problem in timing shift when using the successive adaptation algorithm. This problem will be described below.

FIG. 10 shows the internal structure of the adaptive filter 16 and the so-called FIR is used as the filter section 21.
A specific example using a (finite impulse response) filter is shown. In FIG. 9, the reference input x from the input terminal 16a is the delay elements 23 ₁ , 23 ₂ , ...
..... is sent to a 23 _L series circuit. Input terminal 16
The input x ₀ from a and each delay element 23 ₁ , 23 ₂ , ...
-, 23 each output x ₁ from _L, x _2, ..., x _L, respectively coefficient multipliers 24 _0, 24 _1, 24 _2, ..., is transmitted to the 24 _L, the filter coefficients are w ₀ , W ₁ , w ₂ , ...
., W _L and then sent to the adder 25. Each of the filter coefficients w ₀ , w ₁ , w ₂ , ..., W _L is corrected by the coefficient correction signal from the adaptive algorithm unit 22, and the output y from the adder 25 is taken out from the output terminal 16b.

Many adaptive algorithms have been proposed as the adaptive algorithm used in the adaptive algorithm unit 22. As one specific example thereof, LMS (Least Mean Square, Least Mean), which is one of the successive adaptive algorithms, is proposed.・ Square) algorithm will be explained.

The input data at the k-th sampling period (time k) of the data sequence of the input x and the delayed output data from the delay elements 23 ₁ , 23 ₂ , ..., 23 _L are respectively x _k0 , X _k1 , x _k2 , ..., X _kL , the input vector X _k to be FIR-filtered is X _k = [x _k0 x _k1 x _k2 ... x _kL ] ^T ... (1 ) far. T in the equation (1) represents a transposed symbol. When the respective filter coefficients (weighting coefficients) are w _k0 , w _k1 , w _k2 , ..., W _kL and the FIR filter output is y _k with respect to this input vector X _k , the input / output relationship is as follows. (2)
It becomes like a formula.

Y _k = w _k0 x _k0 + w _k1 x _k1 + ... + w _kL x _kL (2) Further, a filter coefficient vector (weight vector) W
a _k, W _k = if _{[w k0 w k1 w k2 ···} w kL] T ··· (3) and the definition, the input-output ^{_{relationship, y k = X k T ·}} W k ··· (4 ) Is described. If the desired response is d _k , the error ε _{k from the} output is expressed as ε _k = d _k −y _k = d _k −X _k ^T · W _k (5). Using these, the LMS algorithm is expressed as follows: W _{k + 1} = W _k +2 μ · ε _k · X _k (6) In equation (6), μ is a gain factor that determines the speed and stability of adaptation.

When the adaptive filter 16 in which the LMS algorithm is an adaptive algorithm is applied to the configurations shown in FIGS. 7 and 8, it is expressed as shown in FIG. In FIG. 11, reproduction (electric-acoustic conversion) is performed by the speaker 5 of the headphone 1 and sound is collected by the microphone 15 (acoustic-acoustic-
Due to the time delay that occurs before the electrical conversion, the residual error input to the adaptive algorithm unit 22 becomes uncorrelated with the reference input supplied to the filter unit 21 at the same time. Therefore, the condition of the above formula (6) cannot be satisfied. That is, the delay time in the headphones 5 is virtually set to d _1, and the delay time in the microphone 15 is set to d ₂
, The reference input x _k supplied to the filter unit 21
On the other hand, the residual input to the adaptive algorithm unit 22 is delayed by the delay times d ₁ and d ₂ as in e _k-d1-d2 .

As described above, when the timings of the respective inputs supplied to the adaptive filter are deviated, the condition of the above equation (6) is not satisfied, so that effective noise reduction cannot be performed, and the adaptive filter 16 is made to have successive coefficients. There is a drawback that the update type adaptive algorithm cannot be used. In other words, since the amount of calculation is small, the LSI
Therefore, it is not possible to use the line-by-line adaptive algorithm that is suitable for practical use.

For this reason, the applicant of the present invention has previously proposed a noise reduction headphone device which makes it possible to use a so-called LMS algorithm or learning adaptive identification algorithm such as learning identification method in a noise reduction method using an adaptive filter. It was disclosed in Japanese Patent Publication No. 5-30585.

This noise reduction headphone device does not use the circuit having the configuration shown in FIG. 9 as the adaptive filter 16 of the noise reduction headphone device as shown in FIGS. A circuit having the configuration shown is used. Here, each terminal 16a, 16b and 16
Reference characters c correspond to the terminals 16a, 16b and 16c of the adaptive filter 16 as shown in FIG.

In FIG. 12, the reference input x from the terminal 16a is sent to the filter unit 31 of the first adaptive filter 30 and the delay compensation filter 51 described later, and the output from the filter 51 is output. It is sent to the adaptive algorithm unit 32 of the first adaptive filter 30.
The adaptive algorithm unit 32 is supplied with the residual signal e from the terminal 16c.
2 is a filter unit 31 so as to minimize the power of the residual e.
Correct the filter coefficient of. The output y from the filter unit 31 is taken out via the terminal 16b and sent to the filter unit 41 and the adaptive algorithm unit 42 of the second adaptive filter 40, respectively. The output from the filter unit 41 is sent to the adder 52 as a subtraction signal, is subtracted from the residual e from the terminal 16c, and the output from the adder 52 is sent to the adaptive algorithm unit 42. In the adaptive algorithm unit 42, the filter coefficient of the filter unit 41 is sent to the delay compensation filter 51 and copied so as to minimize the power of the output from the adder 52,
The same characteristics (especially delay characteristics) are realized.

Other configurations are the same as those of the noise reduction headphone device shown in FIGS. 7 and 8 and the like, so that they are not shown, but the reference input terminal 16a is provided near the ear when the headphones are attached. The input from the microphone, which is the first acoustic-electric conversion means for collecting external noise (noise), is supplied, and the output from the output terminal 16b is provided near the ear when the headphones are worn. Is sent to the speaker 5 of the headphone 1 which is an electro-acoustic conversion means for outputting sound to the ear canal, and the residual input terminal 1
6c is supplied with an input from a microphone, which is a second acoustic-electrical conversion unit located between the speaker 5 of the headphones 1 and the ear canal when the headphones are attached. Further, the specific internal configuration of each of the first and second adaptive filters 30 and 40 may be the configuration shown in FIG. 9 and FIG. 10 described above, and the filter 51 is the same as the filter 21 of FIG. An FIR filter configuration may be used. Here, the same filter structure is used for the filter unit 41 and the filter 51 of the adaptive filter 40, and the same characteristic (especially delay characteristic) can be realized by copying the filter coefficient.

Next, the operation of the noise reduction headphone device will be described with reference to FIGS. 13 and 14. The operation of the noise reduction headphone device can be roughly divided into an operation centered on the adaptive algorithm section 31 of the adaptive filter 30 and an operation centered on the adaptive algorithm section 41 of the adaptive filter 40. In the following description, it is assumed that the adaptive algorithm unit 41 first performs the adaptive process (FIG. 13), and the adaptive algorithm unit 31 operates after the adaptive process is completed. However, it is also possible to proceed these operations at the same time by appropriately setting the conditions.

In these FIGS. 13 and 14, as in the example of FIG. 8 described above, the sound insulation characteristics of the headphone housing 1 are represented by a transfer function F, which is provided near the ears when the headphones are worn and the external noise ( The transfer function of the microphone 14, which is the first acoustic-electric conversion means provided outside the headphones for picking up noise), is set to M ₁ and is provided near the ears when the headphones are attached to output sound to the ear canal. The transfer function of the speaker 5 of the headphone 1 which is an electro-acoustic conversion means for performing is set to H, and the second sound provided inside the headphone and located between the speaker 5 of the headphone 1 and the ear canal when the headphone is worn- The transfer function of the microphone 15 which is the electrical conversion means is M _2, and the acoustic mixing inside the headphones is shown as addition in the adder 12.

That is, in FIGS. 13 and 14, the external noise (noise) N from the input terminal 11 is electro-acoustically converted by the microphone 14 having the transfer function M ₁ and supplied to the filter section 31 and the filter 51. Along with
The output from the filter unit 31 is electro-acoustic-converted by the speaker 5 of the headphones 1 having the transfer function H and sent to the adder 12. The external noise N is supplied to the adder 12 via the headphone housing 1 having the sound insulation characteristic F and is added and mixed in the headphone inner space to reach the human ear.
The transfer function M ₂ is acoustic-electrically converted by the microphone 15 and sent to the adaptive algorithm unit 32 and the adder 52.
It should be noted that the description of the voice, music signal, etc. originally intended to be reproduced by the headphones (the user intends to listen) does not have a correlation with the noise to be reduced and does not affect the noise reduction operation. is doing.

In FIG. 13, a certain kind of noise is reproduced by the speaker 5 of the headphone 1 and picked up by the microphone 15, and the same kind of noise is also filtered.
And the adaptive algorithm unit 42. The noise picked up by the microphone 14 may be used as this kind of noise, but it is preferable to use the noise generated from a white noise generator or the like. The residual obtained by subtracting the output of the filter unit 41 from the signal picked up by the microphone 15 is input to the adaptive algorithm unit 42. The adaptive algorithm section 42 learns so as to minimize the power of the residual and corrects the coefficient of the filter section magnetic memory chip 41. As a result, the filter unit 41 approximates the characteristics of the headphone 1 through the speaker 5 and the microphone 15, particularly the time delay described above. Each filter coefficient of the filter unit 41 is copied with each filter coefficient of the filter 51. That is, the filter unit 41
The filter 51 has, for example, an FIR filter configuration with the same number of taps, and the same characteristic (delay characteristic) can be realized by copying each coefficient of the filter unit 41 to the filter 51.

Next, referring to FIG. 14, the microphone 14
The noise signal picked up in 1 is filtered by the filter unit 2 to be pseudo noise, and then sent to the speaker 5 of the headphone 1 to be reproduced. This reproduced pseudo noise acoustically cancels noise mixed in the headphones, and the residual thereof is picked up by the microphone 15 and input to the adaptive algorithm unit 32. This residual signal is
Although delayed by the delay in the headphones 1 and the microphone 15, the noise signal supplied as a reference input to the adaptive algorithm unit 32 also has the same delay.

For this reason, the use of a sequential adaptive algorithm such as the LMS algorithm and the learning identification method which requires a small amount of calculation makes it possible to put the LSI into practical use and further facilitates the practical use.

[0029]

By the way, in order for the noise reduction headphone device using the circuit of the configuration shown in FIG. 12 as an adaptive filter to perform a desired operation, the headphone housing 1 covers the user's ear. It is premised that it is worn in close contact with the head.

For example, when the headphone housing 1 is detached from the head or is not in close contact with the head, the coefficient correction operation by the adaptive algorithm is not always performed stably.

There are two types of coefficient correction operation here. First, as described with reference to FIG. 13, the first coefficient correction operation is performed by the adaptive algorithm unit 42 when the power of the residual difference obtained by subtracting the output of the filter unit 41 from the signal picked up by the microphone 15. Train to be the minimum,
This is an operation of correcting the coefficient of the filter unit 41.

The second coefficient correction operation is performed by the adaptive algorithm unit 32 as described with reference to FIG.
This is an operation in which the coefficient of the filter unit 31 is corrected by learning so as to minimize the power of the residual after the acoustic signal output from the driver and the mixed noise are mixed through the filter unit 31.

Further, when the coefficient correction operation of the adaptive filter is continued in a state where the headphone casing 1 is not closely attached to the head and is not worn, convergence for reducing the power of the residual does not occur, and in the worst case, Conversely, divergence occurs in which the power of the residual increases.

If these coefficient correction malfunctions occur, there is a risk that the noise reduction effect will deteriorate or that new noise will be generated by the headphone driver.

The present invention has been made in view of the above circumstances, and it is possible to prevent new noise from being generated by the headphone driver without degrading the noise reduction effect due to the malfunction of the coefficient correction by the adaptive algorithm. An object is to provide a noise reduction headphone device that can be prevented.

[0036]

A noise reduction headphone device according to the present invention is provided with an attachment / detachment detection means for detecting whether or not a headphone housing is attached, and a position near the ear when the headphone is attached, First acoustic-electrical conversion means for collecting external noise, electric-acoustic conversion means for outputting sound to the ear canal provided at a position near the ear when the headphone is mounted, and the electric-acoustic sound when the headphone is mounted. Second acoustic-electrical conversion means located between the conversion means and the ear canal;
Adaptive processing means for adaptively filtering and outputting the input from the first acoustic-electric conversion means, addition means for adding the filtered output from the adaptive processing means and the input voice, and the attachment / detachment detection. The above problem is solved by having an output control means for muting or attenuating the added output supplied from the adding means to the electro-acoustic converting means according to the detected output from the means.

In this case, it is preferable that the adaptive processing means stops the adaptive processing operation according to the detection output from the attachment / detachment detecting means.

The adaptive processing means adaptively filters the input from the first acoustic-electrical converting means so as to minimize the power of the input from the second acoustic-electrical converting means. And a filter unit to which the output from the first adaptive processing unit is supplied, and the output of this filter unit and the output from the second acoustic-electric conversion unit. Second adaptive processing means for adaptively filtering and outputting so as to minimize the power of the residual difference between and.

The attachment / detachment detecting means may detect the attachment / detachment by detecting a mechanical pressing force when the headphone housing is attached to the head, or the first attachment / detachment. Sound-electric conversion means and the second sound-
The attachment / detachment may be detected by comparing the outputs of the electric converting means.

[0040]

[Function] The headphone housing is removed from the head,
When not mounted in close contact, that is, when the coefficient correction operation by the adaptive algorithm is not stable, the coefficient correction operation is stopped and the output of the adaptive filter is turned off. It is possible to prevent the noise reduction effect from being deteriorated due to the malfunction of the coefficient correction by, and to prevent the risk that new noise is generated by the driver of the headphones.

[0041]

Embodiments of the noise reducing headphone device according to the present invention will be described below with reference to the drawings.

First, in the first embodiment, as shown in FIG.
The attachment / detachment detection unit 61 detects whether or not the headphone housing 1 is correctly attached to the head of the user, and when it is correctly attached, the mute circuit or the attenuation circuit 63 is controlled by the output control unit 62, The addition output from the addition circuit 4 is directly supplied to the speaker 5 of the headphone housing 1. On the other hand, when the headphone housing 1 is not properly mounted on the head of the user, the output control unit 63 controls the mute circuit or the attenuation circuit 63 so that the addition output is not directly supplied to the speaker 5.

The adder circuit 4 is supplied with a filter output from the adaptive filter 16 and a voice signal, a music signal or the like via the signal input terminal 7. The adaptive filter 16 includes an input from the first microphone 14 for noise collection, which is provided outside the headphone housing 1 and near the ear when the headphone is worn, and inside the headphone housing 1. At the time of wearing the headphones, the input from the second microphone 15 provided between the speaker 5 of the headphones 1 and the ear canal is supplied.

The adaptive filter 16 filters the input from the first microphone 14 and sends it to the adder 4, and the filter characteristic at this time is set so that the input from the second microphone 15 is minimized. Control adaptively. The internal configuration of the adaptive filter 16 is the same as that in FIG. 9 described above, and the operation thereof can be described with reference to FIG.

The attachment / detachment detection unit 61 detects whether or not the headphone housing 1 is attached in close contact with the head. For example, when the headphone housing is attached to the head, a mechanical detection such as attaching a switch to a part of the headphone housing so as to turn on or off by using the force acting to pinch the head. There is a way. Specifically, (A) of FIG.
As shown in FIG. 2, a movable band portion 73 having a first headphone housing 1a including a speaker unit provided at one end and a second headphone housing 1b including a speaker unit fixed to the other end is provided with a spring. As a structure, a head band 71 movably supported and a fixed switch contact 72a are provided, and the movable portion 7
3 is a method of detecting attachment / detachment by a switching operation between the movable switch contact 72b provided in FIG. FIG. 3B shows a state in which the user wears the noise reduction headphone device. In this case, the fixed switch contact 72a and the movable switch contact 72b may have a structure such as a limit switch or a micro switch. Of course, it is desirable that each switch contact, which is a pressing portion or a protrusion, is not exposed.

Further, for example, the output of a microphone provided outside the housing and the output of a microphone provided inside are compared to obtain a correlation, and when the correlation is high, the wearing state, when the correlation is low, You may make it take the electrical method of determining with a detached state. This method
The principle is that when the headphone device is attached, the noise reaching the internal microphone is smaller than that in the detached state due to the sound insulation effect of the housing. Specifically, the attachment / detachment detection unit 61 may be configured as shown in FIG. That is, the first microphone 14 picks up sound and converts it into an electric signal, and the second microphone 15
Is input and converted into an electric signal by an analog / digital (hereinafter, referred to as A / D) 82,
It is supplied to the correlation coefficient calculator 83 through the A / D 81, and the correlation coefficient calculator 83 takes the correlation. This correlation coefficient is supplied to the comparator 84. The threshold value is supplied to the comparator 84 from the threshold value generator 85. Therefore, the comparator 84 compares whether or not the correlation coefficient from the correlation coefficient calculator 83 is larger than a predetermined threshold value. When the correlation coefficient is larger than the predetermined threshold value, that is, the first microphone 14 and the second microphone 15 are collecting the same sound, and the headphone housing 1
Shows a state in which is detached from the head. On the other hand, when the correlation coefficient is smaller than the predetermined threshold value, that is, the first microphone 14 and the second microphone 15 are collecting different sounds, and the head-on casing 1
Shows a state where is attached to the head. Where the terminal 86
The digital input signal, via 87 and 87, is provided to the adaptive filter 16. Further, the comparison result of the comparator 84 is supplied to the output control unit 62 via the terminal 88.

The output control section 62 receives the detection result from the attachment / detachment detection section 61 and receives the detection result from the mute circuit or attenuation circuit 63.
The output control of the addition output from the addition circuit 4 is performed via. Specifically, when the attachment / detachment detection unit 61 detects attachment / detachment from the head of the headphone housing 1, the addition output is muted or attenuated.

Therefore, the noise reduction headphone device according to the first embodiment is in a state where the headphone housing is not detached from the head or is not mounted in close contact, that is, the coefficient correction operation by the adaptive algorithm is performed. When it is not stable, the output of the adaptive filter is turned off or attenuated, so the noise reduction effect is not deteriorated by the malfunction of coefficient correction by the adaptive algorithm, and new noise is output from the headphone driver. You can prevent the danger that occurs.

Next, a second embodiment will be described with reference to FIG. The noise reduction headphone device of the second embodiment has a block main part as shown in FIG.

In the noise reduction headphone device according to the second embodiment, the attachment / detachment detecting section 61 detects whether or not the headphone housing (not shown) is properly attached to the head of the user. The output control unit 62 turns off the mute circuit or the attenuation circuit 62. On the other hand, when the headphone housing 1 is not properly attached to the head of the user, the output control unit 62 controls the mute circuit or the attenuation circuit 62 to be turned on. Furthermore, the noise reduction headphone device according to the second embodiment has a first adaptive filter 30 and a second adaptive filter 30, which will be described later, according to the detection result of the attachment / detachment detector 61.
The adaptive filter processing of the adaptive filter 40 is controlled by the coefficient correction control unit 64. The structure of the attachment / detachment detector 61 is as shown in FIG. 3 or FIG.

The coefficient correction control unit 64 controls the coefficient correction operation by the adaptive algorithm unit 32 of the first adaptive filter 30 and the coefficient correction operation by the adaptive algorithm unit 42 of the second adaptive filter 40.

There are two types of coefficient correction operation here, as described above. The first coefficient correction operation is the operation shown in FIG.
As described with reference to FIG.
First, the power of the residual difference obtained by subtracting the output of the filter unit 41 from the signal picked up by the microphone 15 is learned, and the coefficient of the filter unit 41 is corrected.

The second coefficient correction operation is the same as in FIG.
As described with reference to FIG. 3, the adaptive algorithm unit 32
First, learning is performed so as to minimize the power of the residual after mixing the acoustic signal output from the driver through the filter unit 31 and the mixed noise, and the coefficient of the filter unit 31 is corrected.

In FIG. 2, the reference input x from the terminal 16a is sent to the filter unit 31 of the first adaptive filter 30 and the delay compensation filter 51 described later, and the output from the filter 51 is the first. Adaptive filter 30
Is sent to the adaptive algorithm unit 32. The residual signal e from the terminal 16c is supplied to the adaptive algorithm unit 32, and the adaptive algorithm unit 32 modifies the filter coefficient of the filter unit 31 so as to minimize the power of the residual e. The output y from the filter unit 31 is controlled by the output control unit 62 and is taken out via the terminal 16b, and at the same time, the filter unit 4 of the second adaptive filter 40 is output.
1 and the adaptive algorithm unit 42 respectively. The output from the filter unit 41 is sent to the adder 52 as a subtraction signal, is subtracted from the residual e from the terminal 16c, and the output from the adder 52 is sent to the adaptive algorithm unit 42. In the adaptive algorithm unit 42, the filter coefficient of the filter unit 41 is set to the delay compensation filter 51 so as to minimize the power of the output from the adder 52.
It is sent to and copied to achieve the same characteristics (especially delay characteristics).

Although not shown in the figure, the other structure is such that the reference input terminal 16a is provided with a first acoustic-electrical device provided near the ear when the headphone is worn and for collecting external noise. An input from a microphone that is a conversion unit is supplied, and an output from the output terminal 16b is a headphone 1 that is an electric-acoustic conversion unit that is provided near the ear when the headphone is worn and that outputs sound to the ear canal. Speaker 5
In addition, the residual input terminal 16c is supplied with an input from a microphone, which is a second acoustic-electric conversion means located between the speaker 5 of the headphones 1 and the ear canal when the headphones are worn. ing. Further, the specific internal configuration of each of the first and second adaptive filters 30 and 40 may be the configuration shown in FIG. 9 and FIG. 10 described above, and the filter 51 is the same as the filter 21 of FIG. An FIR filter configuration may be used. Here, the same filter structure is used for the filter unit 41 and the filter 51 of the adaptive filter 40, and the same characteristic (especially delay characteristic) can be realized by copying the filter coefficient.

The operation of the noise reducing headphone device according to the second embodiment when it is attached to or detached from the user's head will be described below.

First, the case where the user wears the noise reduction headphone device normally will be described. The attachment / detachment detection unit 61 detects that it is in the attached state, and sends a signal indicating this to the coefficient correction control unit 64 and the output control unit 63.

The coefficient correction control section 64 presets the adaptive algorithm section 32 and the adaptive algorithm section 42 and starts the coefficient correction operation. Specifically, as described above, as the first coefficient correction operation, the adaptive algorithm unit 4
2 is learned so that the power of the residual difference obtained by subtracting the output of the filter unit 41 from the signal picked up by the microphone 15 is minimized, and the coefficient of the filter unit 41 is corrected. In addition, as the second coefficient correction operation, the adaptive algorithm unit 3
2 is learned so that the power of the residual after mixing the acoustic signal output from the driver through the filter unit 31 and the mixed noise is minimized, and the coefficient of the filter unit 31 is corrected.

The output control section 63 sends a signal to the mute circuit or the attenuation circuit 62 to allow the output of the filter 31 to pass through without being muted or attenuated.

Next, a case where the user continues to wear the noise reduction headphone device normally will be described. The attachment / detachment detection unit 61 detects that it is in the attached state, and continues to send a signal indicating this to the coefficient correction control unit 64 and the output control unit 63.

The coefficient correction control unit 64 continues the coefficient correction operation of the adaptive algorithm unit 32 and the adaptive algorithm unit 42.

The output control section 63 sends a signal to the mute circuit or attenuator circuit 62 to allow the output of the filter 31 to pass through without muting or attenuating.

Next, a case will be described in which the user has detached the noise reduction headphone device or has not normally worn it.

The attachment / detachment detection section 61 detects that the attachment / detachment state is a detachment state (including a case where the attachment / detachment is not normally performed).
The signal indicating this is sent to the coefficient correction control unit 64 and the output control unit 6
Send to 3.

The coefficient correction control unit 64 stops the coefficient correction operation of the adaptive algorithm unit 32 and the adaptive algorithm unit 42.

The output control unit 63 sends a signal to the mute circuit or the attenuation circuit 62 to mute or attenuate the output of the filter 31.

Next, a case where the noise reduction headphone device is detached or not normally attached will be described.

The attachment / detachment detector 61 detects that the detachment / attachment state (including the case where the attachment / detachment is not normally performed) is continued, and outputs a signal indicating this to the coefficient correction controller 64 and the output controller 63. Keep sending to.

The coefficient correction control unit 64 keeps the coefficient correction operation of the adaptive algorithm unit 32 and the adaptive algorithm unit 42 stopped.

The output control unit 63 sends a signal to the mute circuit or the attenuation circuit 62 to keep the output of the filter 31 muted or attenuated.

Therefore, in the noise reduction headphone device according to the second embodiment, the attachment / detachment detection unit 61 detects attachment / detachment of the headphone case to the head, and the detection result indicates attachment / detachment, that is, the adaptive algorithm. When the coefficient correction operation by is not stable, the coefficient correction operation is stopped and the output of the adaptive filter is turned off or attenuated, so that the noise reduction effect is not deteriorated by the malfunction of the coefficient correction by the adaptive algorithm, Moreover, it is possible to prevent a risk that new noise is generated by the driver of the headphones.

The present invention is not limited to the above embodiment, and the successive adaptive algorithm is not limited to the above LMS method and various other successive adaptive algorithms can be used. As the attachment / detachment detection unit, for example, when the headphone housing is attached to the head,
If you use the force that acts to pinch your head,
The invention is not limited to the specific example shown in FIG.

[0073]

The noise reduction headphone device according to the present invention is provided with the attachment / detachment detection means for detecting whether or not the headphone housing is attached, and the position near the ear when the headphone is attached to collect external noise. A first sound-electricity converting means for making a sound, an electric-acoustic converting means for outputting sound to the ear canal provided at a position near the ear when the headphone is worn, and the electric-acoustic converting means and the ear for wearing the headphone. Second acoustic-electrical converting means located between the road and the road; adaptive processing means for adaptively filtering and outputting the input from the first acoustic-electrical converting means; and the adaptive processing means. An adding means for adding the filtered output and the input sound, and an output for muting or attenuating the added output supplied from the adding means to the electro-acoustic converting means in accordance with the detection output from the attachment / detachment detecting means. In addition to having the control means, the adaptive processing operation of the adaptive processing means can be stopped in accordance with the detection output from the attachment / detachment detecting means, so that the headphone housing is detached from the head or is attached in close contact. When there is no state, that is, when the coefficient correction operation by the adaptive algorithm is not stable, the coefficient correction operation is stopped and the output of the adaptive filter is turned off or attenuated. It is possible to prevent new noise from being generated by the driver of the headphones without deteriorating the noise reduction effect.

[Brief description of drawings]

FIG. 1 is a block circuit diagram showing a configuration of a noise reduction headphone device according to a first embodiment of the present invention.

FIG. 2 is a block circuit diagram showing a configuration of a main part of a noise reduction headphone device according to a second embodiment of the present invention.

FIG. 3 is a schematic configuration diagram of a specific example of a mechanical method of the attachment / detachment detection unit.

FIG. 4 is a block diagram showing a configuration of a specific example of an electrical method of the attachment / detachment detection unit.

FIG. 5 is a configuration diagram of a conventional example of a noise reduction headphone device.

6 is a block diagram for explaining an operation of a conventional example of the noise reduction headphone device shown in FIG.

FIG. 7 is a configuration diagram of another conventional example of a noise reduction headphone device.

8 is a block diagram for explaining the operation of another conventional example of the noise reduction headphone device shown in FIG. 7. FIG.

FIG. 9 is a block diagram of an adaptive filter.

FIG. 10 is a block circuit diagram of a specific example of an adaptive filter.

FIG. 11 is a block diagram for explaining the operation of another conventional example of the noise reduction headphone device.

FIG. 12 is a block diagram of another configuration example of the adaptive filter.

FIG. 13 is an operation explanatory diagram of another configuration example of the adaptive filter.

FIG. 14 is an operation explanatory diagram of another configuration example of the adaptive filter.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Headphone housing 5 Speaker 6 Ear canal 14 First microphone 15 Second microphone 16 Adaptive filter 30 First adaptive filter 31 Filter 32 Adaptive algorithm unit 40 Second adaptive filter 41 Filter 42 Adaptive algorithm unit 61 Attachment / detachment detection unit 62 Mute or Attenuation Circuit 63 Output Control Section

Claims (5)

[Claims] 1. An attachment / detachment detecting means for detecting whether or not a headphone housing is attached, and a first sound-electricity converting means provided at a position near an ear when the headphone is attached and collecting external noise. And an electro-acoustic converting means that is provided near the ear when the headphones are worn and outputs sound to the ear canal, and a second electro-acoustic converting means that is located between the electro-acoustic transducer and the ear canal when wearing the headphones. An acoustic-electric conversion means, an adaptive processing means for adaptively filtering and outputting the input from the first acoustic-electric conversion means, and a filter processing output from the adaptive processing means and an input voice are added. And an output control unit for muting or attenuating the addition output supplied from the addition unit to the electro-acoustic conversion unit according to the detection output from the attachment / detachment detection unit. Noise reduction headphone device. 2. The noise reduction headphone device according to claim 1, wherein the adaptive processing means stops the adaptive processing operation according to a detection output from the attachment / detachment detecting means. 3. The adaptive processing means includes the second sound-
In order to minimize the power of the input from the electrical conversion means,
First adaptive processing means for adaptively filtering and outputting the input from the first acoustic-electric conversion means; and the first adaptive processing means.
Of the adaptive processing means is provided, and the filter is adaptively filtered so as to minimize the power of the residual between the output of the filter portion and the output of the second acoustic-electric conversion means. The noise reduction headphone device according to claim 1, comprising a second adaptive processing means for processing and outputting. 4. The noise reduction headphone according to claim 1, wherein the attachment / detachment detecting means detects attachment / detachment by detecting a mechanical pressing force when the headphone housing is attached to the head. apparatus. 5. The attachment / detachment detection means is configured to detect the first sound.
The noise reduction headphone device according to claim 1, wherein the attachment / detachment is detected by comparing the outputs of the electric converting means and the second acoustic-electric converting means.