Audio earphone with noise reduction device

Abstract

This invention relates to audio headphones with noise reduction devices. The headphones (10) include: a transducer (16); noise reduction processing circuitry (30, 40), including: a microphone (31, 41); and noise reduction processing filters (34, 44), which include, in series: a stabilizing filter (34A, 44A) whose transfer function is substantially equal to the inverse of the transfer function of the processed secondary path; and a noise cancellation filter (34B, 44B) whose transfer function is the noise cancellation transfer function. The secondary path is formed between the transducer (16) and the tympanic membrane, and the transfer function of the processed secondary path is the transfer function of the secondary path affected by the transfer function of components ensuring processing other than the processing filters (34, 44).

Description

[Technical Field]

[0001] This invention relates to noise-canceling casque headphones, the types of which include:

[0002] - An electroacoustic transducer, installed in the sound reproduction cavity;

[0003] - At least one noise reduction circuit, including:

[0004] - A microphone used to capture ambient sounds;

[0005] - Noise reduction filter, used to process the signal from the microphone to generate a noise-reduced signal;

[0006] - A device for applying noise-resistant signals to excite electroacoustic transducers. [Background Technology]

[0007] Noise-canceling headphones include at least one microphone, which is disposed inside or outside a cavity between an electroacoustic transducer and the ear canal. Ideally, such headphones include microphones located in two positions.

[0008] To ensure the generation of an anti-noise signal reproduced by the transducer, the signals from the two microphones are processed by a digital filter, which may be a combination of one or more filters.

[0009] Headphones with built-in microphones generally perform well in terms of noise reduction, typically around 20 to 30 dB. However, this noise reduction is limited to a finite frequency range, typically 50 to 1000 Hz. This is due to the instability of the negative feedback loop formed by the internal microphone and its filters, which directly receive the signal from the electroacoustic transducer. This instability can lead to jitter, so the filters take a series of inferred actions to avoid this phenomenon.

[0010] Headphones with external microphones do not have this instability constraint because the external microphone only picks up about 50dB of already heavily attenuated signal from the electroacoustic transducer, thus avoiding jitter. Headphones with external microphones typically offer attenuation of only 10dB at most because the microphone is biased towards picking up external noise in one pickup direction towards the ear.

[0011] Noise-canceling headphones with external microphones can theoretically attenuate noise at frequencies above 1kHz, but their performance largely depends on the direction of the noise source, especially for high frequencies.

[0012] The filters used at the output of external or internal microphones are designed to overcome the aforementioned problems, namely the jitter of internal microphones and the performance degradation due to strong measurement directivity in the case of external microphones.

[0013] These filters are typically defined empirically to modulate the gain based on the frequency.

[0014] The purpose of this invention is to provide a solution that addresses the problem of internal microphone noise and the difficulty of constructing a high-performance filter for external microphones while considering directionality, while simultaneously enabling a satisfactory widening of the frequency range of the noise reduction device. [Summary of the Invention]

[0015] Therefore, the subject of this invention is noise-canceling headphones of the above type, characterized in that, for this or each noise-canceling processing circuit, the processing filter comprises, in series:

[0016] - A stable filter whose transfer function is essentially equal to the inverse of the transfer function of the processed secondary path, and

[0017] - A noise cancellation filter whose transfer function is the noise cancellation transfer function.

[0018] The secondary path is formed between the electroacoustic transducer and the user's eardrum, and

[0019] The transfer function of the processed secondary path is the transfer function of the secondary path that is affected by the transfer functions of various components in the noise suppression circuit, other than the processing filter, used to ensure processing up to the transducer.

[0020] According to a specific embodiment, noise-canceling headphones include one or more of the following features:

[0021] - The stabilized filter is constructed such that its transfer function is substantially equal to the inverse of the transfer function of the processed secondary path from 5Hz to 50Hz and from 1kHz to 10kHz, with a gain error of about 5dB, and a phase shift of +45 to -45°, said phase being a linear phase corrected for the pure delay due to propagation in the air and the delay due to the processor.

[0022] - It includes internal noise reduction circuitry with an internal microphone housed in the sound reproduction cavity;

[0023] - In the internal noise reduction circuitry, the noise cancellation transfer function has a gain of more than 20dB across the entire audio range;

[0024] -In the internal noise reduction circuit, the stabilizing filter is a proportional-integral filter or a shelf filter.

[0025] - It includes external noise reduction circuitry with an external microphone located outside the sound reproduction cavity;

[0026] - In the external noise cancellation circuit, the noise cancellation transfer function is basically equal to the negative of the quotient of the transfer function corresponding to the passive attenuation of the cavity divided by the transfer function between the outer diaphragm of the cavity and the external microphone.

[0027] The present invention also includes a method for manufacturing noise-canceling headphones, the noise-canceling headphones comprising:

[0028] - Electroacoustic transducer;

[0029] - At least one noise reduction circuit, including:

[0030] - A microphone used to capture ambient sounds;

[0031] - Noise reduction filter, used to process the signal from the microphone to generate a noise-reduced signal;

[0032] - A device for applying noise-resistant signals to excite electroacoustic transducers

[0033] For this or each noise suppression line, the method includes the following steps:

[0034] 1 / Measure the transfer function of the processed secondary path;

[0035] The secondary path is formed between the electroacoustic transducer and the user's eardrum, and

[0036] The transfer function of the processed secondary path is influenced by the transfer functions of various components in the noise suppression circuit, excluding the processing filter, used to ensure processing up to the transducer.

[0037] 2 / Inverse the transfer function of the processed secondary path;

[0038] 3 / Create a processing filter whose transfer function is formed by the following product:

[0039] - The inverse of the transfer function of the processed secondary path, and

[0040] - Noise cancellation transfer function; and

[0041] 4 / Construct headphones whose noise reduction filter is the processing filter created.

[0042] According to a particular implementation, the method includes: measuring the transfer function of the processed secondary path on a complete headphone, but without the noise-canceling filter or each noise-canceling filter, by exciting a transducer with a sinusoidal function of variable frequency over the entire audio range, and measuring the signal obtained on the artificial ear.

Brief Description of the Drawings

[0043] The invention will be better understood by reading the following description, which is given by way of example only and with reference to the accompanying drawings, wherein:

[0044] - Figure 1 This is a schematic diagram of the noise-canceling headphones according to the present invention;

[0045] - Figure 2 The curves, as functions of frequency, show: the gain of the inverse of the transfer function of the processed secondary path, the gain of the stabilizing filter, and the gain of the combination of the transfer function of the processed secondary path and the stabilizing filter.

[0046] - Figure 3 It is a curve, showing the relationship as a function of frequency with... Figure 2 Phases of the same quantity. [Specific implementation method]

[0047] Figure 1 The image shows a noise-canceling headphone 10 schematically.

[0048] It includes a sound reproduction cavity 12, within which the ear 14 of the headphone wearer is schematically shown.

[0049] As is well known, the cavity includes an electroacoustic transducer 16 positioned in the ear canal facing the ear. In the case of external over-ear headphones, the cavity 12 is formed, for example, by a shell covering most of the ear, or in the case of in-ear headphones, by a shell adapted to human anatomy that can be inserted into the entrance of the ear canal.

[0050] To excite transducer 16, transducer 16 is connected to amplifier 18, which is assumed to have unity gain. Amplifier 18 receives the digital signal to be reproduced via digital-to-analog converter 20.

[0051] The headphones include an input 22 for the music signal to be reproduced, which is connected to the input of a digital-to-analog converter 20 via an equalization filter 24.

[0052] To ensure noise reduction, the headphones 10 include an internal noise reduction circuit 30, which includes an internal microphone 31 disposed within the cavity 12 facing the electroacoustic transducer 16.

[0053] The internal microphone 31 is adapted to pick up the sound generated by the transducer 16 and the external noise (denoted as bext) at the outer diaphragm of the cavity 12. The cavity 12 filters the external noise, and the transfer function of the cavity 12 is denoted as HPA.

[0054] The path formed between transducer 16 and the user's tympanic membrane is called the "secondary path" and its transfer function is denoted as Ha.

[0055] The transfer function between the measurement point of the internal microphone 31 and the tympanic membrane is denoted as Hmici-t. Thus, the transfer function between the transducer 16 and the measurement point of the microphone 31 is equal to Ha / Hmici-t.

[0056] In practice, since the distance between the internal microphone 31 and the tympanic membrane is very small, Hmici-t is essentially equal to 1. Therefore, in practice and throughout the rest of this document, both the transfer function of the secondary path and the transfer function between the transducer 16 and the measurement point of the internal microphone 31 are considered to be equal to Ha.

[0057] Microphone 31 is connected in line 30 to an internal signal processing filter 34 suitable for providing noise-canceling signals, in which an analog-to-digital converter 32 is inserted.

[0058] The output of the internal processing filter 34 is connected to the amplifier 18 via an adder 38 located upstream of the digital-to-analog converter 20. This adder ensures the addition of the equalized signal from the input 22 with the noise-resistant signal from the internal processing circuitry 30.

[0059] Similarly, the earphone 10 includes an external noise-canceling circuit 40, which includes an external microphone 41 disposed outside the cavity 12.

[0060] External microphone 41 is adapted to pick up external noise bext affected by the transfer function Hbext. Hbext is the transfer function between the outer surface of cavity 12 where the external noise bext is applied and external microphone 41, such as... Figure 1 As shown.

[0061] In the external processing line 40, the external microphone 41 is connected to the external processing filter 44 via the analog-to-digital converter 42, and the output of the external processing filter 44 is connected to the adder 38.

[0062] In this way, adder 38 ensures that the noise-canceling signal generated at the output of filters 34 and 44 and the equalized music signal to be reproduced from input 22 are sent to amplifier 18 via analog-to-digital converter 20.

[0063] The filters and equalizers described herein are digital filters implemented in a digital signal processor (DSP).

[0064] According to a specific embodiment, the earphone 10 includes two circuits: an internal noise reduction circuit 30 and an external noise reduction circuit 40. Alternatively, the internal noise reduction circuit 30 or the external noise reduction circuit 40 may be removed, and only one of the two associated microphones and filters may be retained.

[0065] According to the present invention, in either embodiment, the external noise reduction filter 34 and the internal noise reduction filter 44, when present, each have a transfer function formed by the following product:

[0066] - The inverse of the transfer function of the processed secondary path, and

[0067] - Noise cancellation transfer function.

[0068] The transfer function of the processed secondary path is influenced by the transfer functions of various components, other than the internal processing filter 34 or the external processing filter 44, used to ensure processing up to the transducer 16, depending on the situation. Here, the transfer function of the processed secondary path specifically relates to the transfer functions of microphone 31 or 41 (depending on the situation), analog-to-digital converter 32 or 42 (depending on the situation), and digital-to-analog converter 20. It is assumed that amplifier 18 is a unity amplifier; if not, the transfer function of amplifier 18 is also incorporated into the transfer function of the processed secondary path.

[0069] The inverse of the transfer function of the processed secondary path is applied to the stable filters 34 and 44, denoted as 34A and 44A, respectively. These filters 34A and 44A have stable transfer functions denoted as HFBcorr and HFFcorr, respectively.

[0070] In the processing filters 34 and 44 respectively, each stabilizing filter 34A and 44A is connected to a noise cancellation filter 34B and 44B after its output. The transfer functions of the noise cancellation filters 34B and 44B are denoted as HFB2 and HFF2 respectively.

[0071] The digital filters used, 34A, 44A and 34B, 44B, are, for example, infinite impulse response (IIR) filters or finite impulse response (FIR) filters.

[0072] The construction and properties of filters 34 and 44 will now be described.

[0073] Let s be the residual noise received by the eardrum of the headphone wearer, assuming it corresponds to the sound picked up by the internal microphone 31.

[0074] The transfer function s / bext without active noise cancellation is denoted as HPA. In other words, it refers to the passive attenuation of the cavity, where bext is the ambient noise on the outer cladding of the cavity.

[0075] The transfer function HPA is typically close to that of a low-pass filter, which means that the structure forming the cavity primarily reduces high frequencies.

[0076] In the Laplace domain, the residual noise is represented by the following expression:

[0077]

Mathematical Formula 1

[0078] s(p)=1 / (1-PlantFB*HFB*exp(-pT FB )*exp(-pD FB ))*(HPA+PlantFF*HFF*exp(-pT FB )*exp(-pD FF )*Hbext)*bext

[0079] in:

[0080] p: complex variable

[0081] HFB: Transfer function of filter 34

[0082] HFF: Transfer function of filter 44

[0083] HPA: Transfer function for passive attenuation of the headphone structure defining cavity 12.

[0084] Hbext: The transfer function between the outer surface of cavity 12 where external noise bext is applied and the external microphone 41.

[0085] PlantFB = Gadci * Gdac * Hmici * Ha is the transfer function of the processed secondary path obtained through the internal noise reduction circuit 30.

[0086] PlantFF = Gadce * Gdac * Hmice * Ha is the transfer function of the processed secondary path obtained through the external noise reduction circuit 40.

[0087] in:

[0088] Gadci and Gadce: Gains of analog-to-digital converters 32 and 42 used for internal microphone 31 and external microphone 41, respectively.

[0089] Gdac: Output gain of digital-to-analog converter 20

[0090] hmici: Transfer function of internal microphone 31

[0091] Hmice: Transfer function of external microphone 41

[0092] Ha: Assuming the transfer function between the tympanic membrane and transducer 16 corresponding to the measurement point of the internal microphone 31.

[0093] Ha depends on the characteristics of the transducer and the acoustic architecture around it, especially the acoustic architecture of the chambers in front of and behind the transducer when the transducer is a loudspeaker.

[0094] Ha represents the transfer function of the secondary path, that is, the transfer function between the transducer 16 and the positioning point or tympanic membrane of the internal microphone 31 without considering delay. In modeling, the propagation time of the sound wave is isolated within a specific term exp(-pT). FB In this way, taking this delay into account, Hareal = Ha * exp(-pT) FB The complete transfer function Hareal is represented by ).

[0095] T FB The distance d between the transducer 16 and the internal microphone 31 for sound waves FB The propagation time on T FB =d FB / c, where c is the speed of sound (342 m / s)

[0096] D FB Processing delay between the input and output terminals of the digital signal processor with internal noise reduction circuitry 30

[0097] D FF : Processing delay between the input and output terminals of the digital signal processor in the external noise reduction circuit 40.

[0098] Now consider the first embodiment, in which the external noise reduction circuit 40 is removed.

[0099] In this case, the residual noise at the tympanic membrane is represented as follows:

[0100]

Mathematical Formula 2

[0101] s(p)=1 / (1–PlantFB*HFB*exp(-pT FB )*exp(-pD FB ))*(HPA)*bext

[0102] It is measured and processed solely from the internal microphone 31.

[0103] Since filter 34 is formed by two parts, namely noise cancellation filter 34B with transfer function denoted as HFB2 and stable filter 34A with transfer function HFBcorr, we have: HFB = HFBcorr * HFB2.

[0104] According to the present invention, the transfer function HFBcorr is considered to be essentially equal to the inverse of the transfer function PlantFB of the processed secondary path, that is:

[0105]

Mathematical Expression 3

[0106] HFBcorr*PlantFB ~=1

[0107] Therefore, the stabilizing filter 34A is constructed by essentially applying the transfer function HFBcorr ~ = PlantFB. -1 .

[0108] To construct filter 34A, firstly, the transfer function PlantFB of the processed secondary path is measured on an earphone that is fully but not programmed internally into a digital signal processor (DSP) to excite transducer 16 with a sine function of variable frequency across the entire audio range, and the signal obtained by artificial ear 14 is measured, which allows the value of Ha to be determined. The values ​​of other terms are known, i.e., the transfer functions of commercially available components.

[0109] Next, the inverse PlantFB of this transfer function will be calculated numerically. -1 .

[0110] Because the transfer function PlantFB does not incorporate the delay exp(-pT) FB ) and exp(-pD FB Therefore, the transfer function is invertible, and its causal relationship can be reversed by a filter in a real-time system. For this reason, delay is not considered in the transfer function of the processed secondary path. Otherwise, although the terms constituting the delay are invertible, it would be impossible to construct a filter that implements its inverse in a real-time system, because the filter would then become anti-causal.

[0111] Therefore, a stable filter 34A was constructed, which reproduces the transfer function PlantFB as faithfully as possible. -1 The transfer function HFBcorr.

[0112] In practice, the stabilizing filter 34A is constructed such that its transfer function is essentially equal to the inverse of the transfer function of the secondary path over the entire audio range, and especially from 5Hz to 50Hz and from 1kHz to 10kHz, with a gain error of about 5dB, advantageously 1dB, and a phase shift of +45 to -45°, said phase being a linear phase corrected for the pure delay due to propagation in the air and for the delay due to the processor.

[0113] The filter is programmed and implemented in a digital signal processor (DSP). It is advantageously implemented by combining several cascaded filters.

[0114] The stabilizing filter 34A operates across the entire frequency range allowed by the sampling frequency (Fs) of the digital signal processor (DSP). For example, if Fs = 384 kHz, then the correction range of filter 34A is 0 Hz to 192 kHz.

[0115] The second part of filter 34 consists of a noise cancellation filter 34B with a transfer function HFB2. This second part of filter 34 is designed to ensure stability across all frequencies while applying the highest possible gain in the audio band, particularly greater than 20 dB, to provide maximum noise cancellation performance. Filter 34B is advantageously formed by a proportional-integral (PI) filter or by a shelving filter.

[0116] Only the residual noise of the internal noise reduction circuitry 30 of the internal microphone 31 is:

[0117] s(p)=1 / (1-PlantFB*HFB*exp(-pT FB )*exp(-pD FB ))*(HPA)*bext

[0118]

Mathematical Formula 2

[0119] By incorporating [Mathematical Expression 3] into [Mathematical Expression 2], we obtain:

[0120] s(p)~=1 / (1-HFB2*exp(-pT FB )*exp(-pD FB ))*(HPA)*bext.

[0121] Understandably, the phase of the denominator depends only on the pure delay, i.e., the physical propagation time T of the sound wave. FB Processing time D of a digital signal processor (DSP) FB The sum of these factors, along with the phase of the noise cancellation filter HFB2 which excludes denominator cancellation in the frequency band, is the source of jitter. Thus, this situation can be avoided even with high gain noise immunity over a wide frequency range.

[0122] Figure 2 and Figure 3 The results are shown in the figure.

[0123] In these graphs, the inverse of the transfer function, PlantFB -1 It is represented by a thin solid line. It is the exact mathematical inverse of the transfer function PlantFB, which was experimentally measured on headphones.

[0124] The actual transfer function HFBcorr of the implemented stable filter 34A is shown as a dotted line. As mentioned earlier, the two curves are very close.

[0125] Thus, the difference between PlantFB*HFBcorr, represented by the thick solid line, corresponds to the gain ( Figure 2) and phase ( Figure 3 The transfer function is actually flat in terms of the high-frequency phase difference, which is mainly due to the pure delay of the system, if the following is considered:

[0126] D FB =11μs

[0127] T FB ~ = 6μs.

[0128] Now consider the second embodiment, in which the internal noise reduction circuit 30 is removed and only the external noise reduction circuit 40 exists.

[0129] In this case, the residual noise at the tympanic membrane is represented as follows:

[0130]

Mathematical Expression 4

[0131] s(p)=(HPA+PlantFF*HFF*exp(-pTFB)*exp(-pDFF)*Hbext)*bext

[0132] It is measured and processed solely from the external microphone 41.

[0133] Similarly, for the external noise reduction processing line 40, the noise cancellation filter 44B with a transfer function denoted as HFF2 and the stabilizing filter 44A with a transfer function HFFcorr, we have: HFF = HFFcorr * HFF2.

[0134] According to the present invention, the transfer function HFFcorr is considered to be essentially equal to the inverse of PlantFF, such that:

[0135]

Mathematical Expression 5

[0136] FFcorr*PlantFF~=1, that is, HFFcorr~=PlantFF -1 .

[0137] In this case, the processed secondary path is given by the following formula:

[0138] PlantFF=Gadce*Gdac*Hmice*Ha.

[0139] To construct filter 44A, firstly, the transfer function PlantFF of the processed secondary path is measured on a full headphone without internal processing filter 34 by subjecting transducer 16 to a variable frequency that scans the audio range, and the signal obtained by artificial ear 14 is measured.

[0140] Next, the inverse PlantFF of this transfer function is calculated numerically. -1 .

[0141] First, the PlantFF transfer function of the transducer 16 was measured on the complete earphone without the internal processing filter 44, through a secondary path that subjected the transducer 16 to variable scan frequency processing. The range and measurement of the signal obtained by the artificial ear are then discussed.

[0142] Then the inverse PlantFF-1 of the transfer function is calculated numerically.

[0143] In practice, as in the previous embodiment, the stabilizing filter 44A is configured such that its transfer function is substantially equal to the inverse of the transfer function of the secondary path from 5 Hz to 50 Hz and from 1 kHz to 10 kHz, and advantageously is substantially equal to the inverse of the transfer function of the secondary path over the entire audio range, with a gain error of about 5 dB, advantageously 1 dB, and a phase shift of +45 to -45° in phase, without considering linear phase and delay due to processor delay, the linear phase being the pure delay caused by propagation in the air.

[0144] The residual noise is for a single external noise reduction circuit 40 with only external microphone 41, and can be written in the following form:

[0145] s(p)=(HPA+PlantFF*HFF*exp(-pT FB )*exp(-pD FF )*Hbext)*bext

[0146]

Mathematical Expression 4

[0147] By incorporating [Mathematical Expression 5] into [Mathematical Expression 4], we obtain:

[0148] s(p)=(HPA+HFF2*exp(-pT FB )*exp(-pD FF )*Hbext)*bext.

[0149] This form allows for the definition of filter HFF2 without considering constraints on other components of the system, which in the above expression are simplified to a simple delay exp(-pT). FF )*exp(-pD FF ).

[0150] The optimal transfer function of filter 44, denoted as HFFopt, corresponding to the cancellation of residual noise as expressed in [Mathematical Equation 4], is as follows:

[0151] HFFopt=-(HPA / Hbext) / (PlantFF)*exp(-pT FB )*exp(-pD FF ).

[0152] Choose noise cancellation filter 44B, making its transfer function equal to HFF2 = -HPA / Hbext, so that the residual noise is:

[0153] s(p)=HPA*(1-exp(-pT FB )*exp(-pD FF ))*bext.

[0154] Due to delay T FB and D FF The product of the two exponents is very small, therefore -exp(-pT) FB )*exp(-pD FF The value approaches 1 over a wide frequency range, such that (1-exp(-pT)) FB )*exp(-pD FF The value is very close to zero, which corresponds to very strong noise attenuation over a wide frequency range.

[0155] Now consider the third embodiment, in which both internal noise reduction circuitry 30 and external noise reduction circuitry 40 are present.

[0156] In the Laplace domain, the residual noise is represented by the following expression:

[0157] s(p)=1 / (1-PlantFB*HFB*exp(-pT FB )*exp(-pD FB ))*(HPA+PlantFF

[0158] *HFF*exp(-pT FB )*exp(-pD FF )*Hbext)*bext

Mathematical Expression 1

[0159] In this case, as mentioned earlier, the transfer functions of processing filters 34 and 44 are expressed in the following form:

[0160] HFB = HFBcorr * HFB2, where HFBcorr is constructed such that HFBcorr * PlantFB ~ = 1, and

[0161] HFF = HFFcorr * HFF2, where HFFcorr is constructed such that HFFcorr * PlantFF ~ = 1.

[0162] The residual noise is therefore written as:

[0163]

Mathematical Expression 6

[0164] s(p)=1 / (1-HFB2*exp(-pTFB )*exp(-pD FB ))*(HPA+HFF2*exp(-pT FB )*exp(-pD FF )*Hbext)*bext.

[0165] Furthermore, advantageously, by choosing HFF2 = -HPA / Hbext as described above, we have:

[0166] s(p)=1 / (1-HFB2*exp(-pT FB )*exp(-pD FB ))*HPA*(1-exp(-pT FB )*exp(-pD FF ))*bext.

[0167] This embodiment combines the advantages of the previous two embodiments.

Claims

1. Noise-canceling headphones (10), comprising: - An electroacoustic transducer (16) is installed in the sound regeneration cavity (12); - At least one noise reduction circuit (30, 40), including: - Microphones (31, 41) used to capture ambient sounds; - Noise reduction filters (34, 44) are used to perform noise reduction processing on the signals from the microphones (31, 41) to generate noise-reduced signals; - Devices (18, 20) for applying noise-resistant signals to excite the electroacoustic transducer (16) For each noise reduction processing line (30, 40), the processing filters (34, 44) include those connected in series: - Stabilized filters (34A, 44A) whose transfer function is equal to the inverse of the transfer function (HFBcorr, HFFcorr) of the processed secondary paths, and - Noise cancellation filters (34B, 44B), whose transfer functions are noise cancellation transfer functions (HFB2, HFF2). The secondary path is formed between the electroacoustic transducer (16) and the user's eardrum, and The transfer function of the processed secondary path is the transfer function of the secondary path that is affected by the transfer functions of various components in the noise suppression processing lines (30, 40) other than the processing filters (34, 44) used to ensure processing up to the transducer (16). Its characteristic feature is that it includes an external noise reduction circuit (40) having an external microphone (41) disposed outside the sound reproduction cavity (12); and In the external noise cancellation circuit (40), the noise cancellation transfer function (HFF2) is equal to the negative of the quotient of the transfer function corresponding to the passive attenuation of the cavity (12) divided by the transfer function between the outer diaphragm of the cavity (12) and the external microphone (41).

2. The noise-canceling headphones (10) according to claim 1, characterized in that, The stabilizing filters (34A, 44A) are constructed such that their transfer functions (HFBcorr, HFFcorr) from 5Hz to 50Hz and from 1kHz to 10kHz are equal to the inverse of the transfer functions (PlantFB, PlantFF) of the processed secondary paths, with a gain error of 5dB, and a phase shift of +45 to -45°, said phase being a linear phase corrected for the pure delay due to propagation in the air and the delay due to the processor.

3. The noise-canceling headphones (10) according to claim 1 or 2, characterized in that, It includes an internal noise reduction circuit (30) with an internal microphone (31) housed in the sound reproduction cavity (12).

4. The noise-canceling headphones (10) according to claim 3, characterized in that, In the internal noise reduction circuitry (30), the noise cancellation transfer function (HFB2) has a gain of more than 20 dB over the entire audio range.

5. The noise-canceling headphones (10) according to claim 3 or 4, characterized in that, In the internal noise reduction circuit (30), the stabilizing filter (34A) is a proportional-integral filter or a shelf filter.

Images