JP2016218456A - Noise reducing sound reproduction - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a noise reducing sound reproduction system.SOLUTION: A noise reducing sound reproduction system includes: a loudspeaker 3 that is connected to a loudspeaker input path; a microphone 4 that is acoustically coupled to the loudspeaker via a secondary path 2 and connected to a microphone output path; a first subtractor 8 that is connected downstream of the microphone output path and a first useful-signal path; an active noise reduction filter 5 that is connected downstream of the first subtractor; and a second subtractor 9 that is connected between the active noise reduction filter and the loudspeaker input path and to a second useful-signal path. Two useful-signal paths are transmitted with a useful signal to be reproduced, and at least one of the two useful-signal paths comprises one or more spectrum shaping filters 10.SELECTED DRAWING: Figure 4

Description

  The present invention relates to a noise-removed audio reproduction system, and more particularly to a noise removal system including an earphone that allows a user to enjoy reproduced music or the like by removing ambient noise.

  In an active noise cancellation system, also known as an active noise cancellation / control (ANC) system, the same speaker, specifically a speaker placed inside the two earphones of a headphone, is often used for noise reduction and music and speech, etc. Used for both playing back the audio you want to listen to. However, due to the fact that common denoising systems reduce the desired audio as much as possible, there is a significant difference in the impression of audio when using and not using an active denoising system. For this reason, to compensate for such effects, advanced electrical signal processing is required, or listening to the impression of audio that is affected by turning noise removal on or off. The hands needed to be accepted.

  Accordingly, there is a general need for an improved noise removal system that overcomes the above disadvantages.

  A first aspect of the present invention includes a speaker connected to a speaker input path, a microphone acoustically coupled to the speaker via a secondary path and connected to a microphone output path, a microphone output path, A first subtractor coupled downstream of one useful signal path, an active noise cancellation filter coupled downstream of the first subtractor, and coupling between the active noise cancellation filter and the speaker input path And a second subtractor coupled to the second useful signal path, wherein the regenerated useful signal is transmitted to both useful signal paths, at least one of the useful signal paths being one This is a noise-removed sound reproduction system comprising the above spectrum shaping filter.

  According to a second aspect of the present invention, an input signal acoustically radiated by a speaker is transmitted to the speaker, and a microphone acoustically coupled to the speaker via a secondary path and transmitting a microphone output signal is a signal radiated by the speaker. Receives and subtracts useful signal from microphone output signal to generate filter input signal, filters filter input signal with active noise removal filter to generate error signal, subtracts useful signal from error signal and speaker input A denoising speech reproduction method comprising generating a signal and filtering the useful signal with one or more spectral shaping filters before subtracting the useful signal from the microphone output signal or the speaker input signal or both signals.

  Various specific embodiments are described in detail below based on the embodiments illustrated in the drawings. Unless otherwise stated, like or identical components are provided with like reference numerals throughout all of the drawings.

For example, the present invention provides the following items.
(Item 1)
A noise-removal audio playback system,
A speaker connected to the speaker input path;
A microphone that is acoustically coupled to the speaker via a secondary path and coupled to a microphone output path;
A first subtractor coupled downstream of the microphone output path and a first useful signal path;
An active noise removal filter coupled downstream of the first subtractor;
A second subtractor coupled between the active noise removal filter and the speaker input path and coupled to a second useful signal path;
Including
A useful signal to be reproduced is sent to the two useful signal paths,
At least one of the two useful signal paths includes one or more spectral shaping filters;
Noise-removal audio playback system.
(Item 2)
The secondary path has a secondary path transfer characteristic, and at least one of the spectral shaping filters models the secondary path transfer characteristic or linearizes a microphone signal on the microphone output path with respect to the useful signal The system according to the above item, wherein the system has transfer characteristics.
(Item 3)
The system according to any of the preceding items, wherein the first useful signal path includes a first spectral shaping filter having a transfer characteristic that matches a transfer characteristic of the secondary path.
(Item 4)
The system according to any of the preceding items, wherein the first spectral shaping filter comprises at least two sub-filters.
(Item 5)
The system according to any of the preceding items, wherein at least one of the first spectral shaping filter or the sub-filter of the first spectral shaping filter is an equivalent filter.
(Item 6)
The system according to any of the preceding items, wherein the equivalent filter is a treble cut equivalent filter.
(Item 7)
The system according to any of the preceding items, wherein one of the first spectral shaping filter or the sub-filter of the first spectral shaping filter is a shelving filter.
(Item 8)
The system according to any of the preceding items, wherein the shelving filter is a treble cut shelving filter.
(Item 9)
A system according to any of the preceding items, comprising a second spectral shaping filter, wherein the second useful signal path has a transfer characteristic that matches the reverse transfer characteristic of the secondary path.
(Item 10)
The system according to any of the preceding items, wherein at least one of the active noise removal filter, the first spectral shaping filter, and the second spectral shaping filter is an adaptive filter.
(Item 11)
A noise-removal audio playback method,
Transmitting an input signal radiated acoustically by the speaker to the speaker;
Receiving the signal emitted by the speaker by a microphone that acoustically couples with the speaker via a secondary path and transmits a microphone output signal;
Subtracting the microphone output signal from a useful signal to generate a filter input signal;
Filtering the filter input signal with an active noise removal filter to generate an error signal;
Subtracting the useful signal from the error signal to generate the speaker input signal;
Filtering the useful signal with one or more spectral shaping filters before subtracting the useful signal from the microphone output signal or the speaker input signal or both signals;
Including methods.
(Item 12)
A method as described above, wherein the secondary path has secondary path transfer characteristics and all spectrum shaping filters model the transfer characteristics of the secondary path as a whole.
(Item 13)
A method according to any of the preceding items, wherein the useful signal is filtered by a transfer characteristic that matches the transfer characteristic of the secondary path before subtracting the useful signal from the microphone output signal.
(Item 14)
The method according to any of the preceding items, wherein filtering of the useful signal comprises equivalent and / or shelving filtering.
(Item 15)
A method according to any of the preceding items, wherein the useful signal is filtered by a transfer characteristic that matches the reverse secondary path transfer characteristic before subtracting the useful signal from the speaker input signal.

(Summary)
The present invention transmits an input signal acoustically radiated by a speaker to the speaker, and acoustically couples with the speaker via a secondary path and transmits a microphone output signal. Is subtracted from the microphone output signal to generate a filter input signal, the filter input signal is filtered by an active noise removal filter to generate an error signal, and a useful signal is subtracted from the error signal to generate a speaker input signal. A denoising audio reproduction system and method is disclosed that includes filtering the useful signal with one or more spectral shaping filters before subtracting the useful signal from the microphone output signal or speaker input signal or both. .

1 is a block diagram illustrating a general feedback active noise removal system in which useful signals are transmitted to a signal path of a speaker. 1 is a block diagram illustrating a general feedback active noise removal system in which useful signals are transmitted to a microphone signal path. FIG. 1 is a block diagram illustrating a general feedback active noise removal system in which useful signals are transmitted to a speaker and microphone signal path. FIG. FIG. 4 is a block diagram illustrating the active noise removal system of FIG. 3 in which useful signals are transmitted via a spectral shaping filter to a speaker path. FIG. 4 is a block diagram illustrating the active noise removal system of FIG. 3 in which useful signals are transmitted to the microphone path via a spectral shaping filter. It is a figure which shows schematically the earphone which can be applied to the connection with the active noise removal system of FIGS. FIG. 6 is a block diagram illustrating the active noise removal system of FIG. 5 in which useful signals are transmitted to the microphone path via two spectral shaping filters. It is an amplitude frequency response figure which shows the transfer characteristic of the shelving filter applicable in the system of FIG. It is an amplitude frequency response figure which shows the transfer characteristic of the equivalent filter applicable in the system of FIG. FIG. 6 is a block diagram illustrating the active noise removal system of FIG. 5 in which useful signals are transmitted to both the microphone and speaker paths via a spectral shaping filter.

  The feedback ANC system removes or cancels interference signals such as noise by providing a noise removal signal at the listening site that is ideally the same amplitude over time but has an opposite phase compared to the noise signal. Is provided for. A signal obtained by superimposing a noise signal and a noise removal signal (also known as an “error signal”) ideally tends to zero. The quality of the denoising is influenced by the so-called secondary path, ie the quality of the acoustic path between the speaker and the microphone instead of the listener's ear. The quality of noise removal is further influenced by the quality of the so-called ANC filter. The ANC filter is coupled between the microphone and the speaker and serves to filter the error signal provided by the microphone. The ANC filter also acts to further remove the error signal when the filtered error signal is reproduced by a speaker. However, problems arise particularly when useful signals, such as music and speech, are additionally provided to the filtered error signal at the listening site by speakers that also reproduce the filtered error signal. Thus, this useful signal can be degraded by the system described above.

  For simplicity, no distinction is made between electrical and acoustic signals in this specification. However, all signals transmitted by the speaker and received by the microphone actually contain acoustic properties. All other signals include electrical properties. The speaker and microphone are part of an acoustic subsystem (eg, a speaker room microphone system) having an input stage formed by the speaker 3 and an output stage formed by the microphone, the electrical input to this subsystem A signal is transmitted and the subsystem transmits an electrical output signal. In this regard, a “path” is an electrical or acoustic connection that may include additional elements such as, for example, signal conducting means, amplifiers, filters, and the like. The spectrum shaping filter is a filter in which the spectrums of the input signal and the output signal differ depending on the frequency.

  Reference is now made to FIG. 1, which is a block diagram illustrating a general feedback active noise removal (ANC) system. In this feedback-type active noise removal system, the interference signal d [n] (also referred to as a noise signal) is transmitted (radiated) to the listening site, for example, the listener's ear, via the primary path 1. The primary path 1 has a transfer characteristic of P (z). Further, the input signal v [n] is transmitted (radiated) from the speaker 3 to the listening site via the secondary path 2. The secondary path 2 has a transfer characteristic S (z).

  The microphone 4 arranged at the listening site receives the signal generated from the speaker 3 together with the disturbing signal d [n]. Microphone 4 provides a microphone output signal y [n] representing the sum of these received signals. The microphone output signal y [n] is transmitted to the ANC filter 5 as a filter input signal u [n], and the ANC filter 5 outputs an error signal e [n] to the adder 6. The ANC filter 5 may be an adaptive filter and has a transfer characteristic W (z). The adder 6 also receives a useful signal x [n] such as music or speech prefiltered as necessary by a spectrum shaping filter (not shown) and transmits the input signal v [n] to the speaker 3. To do.

  The signals x [n], y [n], e [n], u [n] and v [n] are in the discrete time domain. These spectral representations, X (z), Y (z), E (z), U (z) and V (z) are used for the following considerations. The differential equation describing the system shown in FIG.

図１のシステムにおいて、有用信号の伝達特性 Ｍ（ｚ）＝Ｙ（ｚ）／Ｘ（ｚ）はしたがって、

In the system of FIG. 1, the transfer characteristic of useful signal M (z) = Y (z) / X (z) is therefore

である。

It is.

  As can be seen from equations (4) to (7), the transfer characteristic M (z) of the useful signal approaches zero (0) as the transfer characteristic W (z) of the ANC filter 5 increases. The function S (z) remains near neutral level 1, that is, 0 [dB]. For this reason, it is necessary to appropriately adapt the useful signal x [n] so that the listener can reliably capture the useful signal x [n] when the ANC is turned on or off. Further, the transfer characteristic M (z) of the useful signal is also influenced by the transfer characteristic S (z) of the secondary path 2, and the adaptation of the useful signal x [n] is also determined by the transfer characteristic S (z), its time, temperature, listener A certain difference between “on” and “off” is clarified because it is affected by fluctuations due to changes of the

  In the system of FIG. 1, the useful signal x [n] is transmitted to the acoustic subsystem (speaker, room, microphone) of the adder 6 connected upstream of the speaker 3, but in the system of FIG. 2, the useful signal x [N] is transmitted to the microphone 4. Thus, in the system of FIG. 2, the adder 6 is omitted, and the adder 7 is arranged downstream of the microphone 4, for example, a prefiltered useful signal x [n] and a microphone output signal y [n]. Find the sum of Therefore, the speaker input signal v [n] is equal to the error signal [e], that is, v [n] = [e], and the filter input signal u [n] is the useful signal x [n] and the microphone output signal y. The sum with [n], that is, u [n] = x [n] + y [n].

  The differential equation describing the system shown in FIG. 2 is as follows:

図２のシステムにおいて、妨害信号ｄ［ｎ］を考慮しなかった場合の有用信号の伝達特性Ｍ（ｚ）は以下の通りである：

In the system of FIG. 2, the transfer characteristic M (z) of the useful signal without considering the interference signal d [n] is as follows:

式（１１）〜（１３）から分かるように、有用信号の伝達特性Ｍ（ｚ）は、開ループ伝達特性（Ｗ（ｚ）・Ｓ（ｚ））が増加または減少すると、１に近づき、開ループ伝達特性（Ｗ（ｚ）・Ｓ（ｚ））が０（ゼロ）に近づくと、０（ゼロ）に近づく。このため、有用信号ｘ［ｎ］は、ＡＮＣのオンまたはオフ時に聴き手が有用信号ｘ［ｎ］をまったく同じに確実に捕えることができるように更に高いスペクトル範囲に適応する必要がある。しかしながら、高いスペクトル範囲の補償はきわめて困難であるため、オンとオフの間である一定の差異が明確になる。一方、有用な伝達特性Ｍ（ｚ）は、二次経路２の伝達特性Ｓ（ｚ）とその経時、温度、聴き手の変化などによる変動とに影響されない。

As can be seen from the equations (11) to (13), the transfer characteristic M (z) of the useful signal approaches 1 when the open loop transfer characteristic (W (z) · S (z)) increases or decreases, and opens. When the loop transfer characteristic (W (z) · S (z)) approaches 0 (zero), it approaches 0 (zero). Therefore, the useful signal x [n] needs to be adapted to a higher spectral range so that the listener can capture the useful signal x [n] exactly the same when the ANC is on or off. However, compensation for high spectral range is very difficult, so certain differences between on and off become clear. On the other hand, the useful transfer characteristic M (z) is not affected by the transfer characteristic S (z) of the secondary path 2 and its variation with time, temperature, listener, and the like.

  FIG. 3 is a block diagram showing a general feedback active noise removal system in which useful signals are transmitted to both the speaker and microphone paths. For the sake of simplicity, the primary path 1 still has noise (disturbance signal d [n]), but is omitted below. Specifically, the system of FIG. 3 further includes a subtractor 8 that subtracts the useful signal x [n] from the microphone output signal y [n] to form the ANC filter input signal u [n], and the adder 6 1 except that a subtracter 9 for subtracting the useful signal x [n] from the error signal e [n] is provided.

  A differential equation for explaining the system shown in FIG. 3 is expressed as follows.

図３のシステムにおける有用信号の伝達特性Ｍ（ｚ）は以下のように表される。

The useful signal transfer characteristic M (z) in the system of FIG. 3 is expressed as follows.

式（１７）〜（１９）から、図３のシステムの振る舞いは図２のシステムと同様であることが判断できる。唯一の違う点は、開ループ伝達特性（Ｗ（ｘ）・Ｓ（ｚ））がゼロ（０）に近づくと、有用信号の伝達特性Ｍ（ｚ）がＳ（ｚ）に近づくことである。図１のシステムと同様に、図３のシステムは、二次経路２の伝達特性Ｓ（ｚ）とその経時、温度、聴き手の変化などに因る変動とに影響される。

From the equations (17) to (19), it can be determined that the behavior of the system of FIG. 3 is the same as that of the system of FIG. The only difference is that the useful signal transfer characteristic M (z) approaches S (z) as the open loop transfer characteristic (W (x) · S (z)) approaches zero (0). Similar to the system of FIG. 1, the system of FIG. 3 is affected by the transfer characteristic S (z) of the secondary path 2 and its variation over time, temperature, listener changes, and the like.

  In FIG. 4, an equivalent filter 10 based on the system of FIG. 3 and connected upstream of the subtractor 9 to filter the useful signal x [n] by the inverse secondary path transfer function 1 / S (z). A system including further is shown. The differential equations describing the system shown in FIG.

図４のシステムにおける有用信号の伝達特性Ｍ（ｚ）は従って、

The useful signal transfer characteristic M (z) in the system of FIG.

となる。

It becomes.

  As can be seen from equation (22), the microphone output signal y [n] is identical to the useful signal x [n]. That is, if the equivalent filter is a true reciprocal of the secondary path transfer characteristic S (z), the signal x [n] is not changed by the system. This equivalent filter 10 is the minimum phase filter for the optimal result, i.e. ideally the minimum phase, the optimal approximation of its actual transfer characteristic towards the inverse of the transfer characteristic S (z) of the second path, and therefore , Y [n] = x [n]. This configuration acts as an ideal linearizer. That is, this configuration compensates for the deterioration of the useful signal caused by the transmission from the speaker 3 to the microphone 4 that is the ear of the listener. Thus, this configuration compensates or linearizes the disturbing effect on the useful signal x [n] in the secondary path S (z), thereby providing the sound source without negative effects due to the acoustic characteristics of the headphones. A useful street signal arrives at the listener. That is, y [z] = x [z]. In this manner, with such a linear filter, even a poorly designed headphone can generate acoustically perfectly adjusted sound, that is, linear sound.

  In FIG. 5, a system based on the system of FIG. 3 and further comprising an equivalent filter 10 coupled upstream of the subtractor 8 to filter the useful signal x [n] by the secondary path transfer function S (z). It is shown. The differential equations describing the system shown in FIG.

となる。

It becomes.

  Judging from the equation (25), it can be seen that when the ANC system is active, the transfer characteristic M (z) of the useful signal matches the secondary path transfer characteristic S (z). Even when the ANC system is not active, the transfer characteristic M (z) of the useful signal matches the secondary path transfer characteristic S (z). Thus, the auditory impression when a useful signal for the listener near the microphone 4 reaches the ear is the same regardless of whether or not noise removal is active.

  The ANC filter 5 and the equivalent filters 10 and 11 are each a fixed filter having a fixed transfer characteristic or an adaptive filter having an adjustable transfer characteristic. In the drawing, the adaptive structure of its own filter is indicated by an arrow below each block, and the selectivity of the adaptive structure is indicated by a broken line.

  The system shown in FIG. 5 is applicable to headphones in which useful signals such as music and voice are reproduced under different conditions with respect to noise, and in particular, in the absence of noise, the listener can use the ANC system. It will be appreciated that it is possible to switch off the ANC system without experiencing the difference between the active and inactive sounds. However, the system presented herein is applicable not only to headphones, but also to all other fields where accidental noise removal is desired.

  FIG. 6 illustrates an earphone in which the active noise removal system of the present invention may be used. The earphone, together with another similar earphone, forms part of a headphone (not shown) and is acoustically coupled to the listener's ear 12. In an embodiment of the invention, the ear 12 is exposed to the disturbing signal d [n], for example ambient noise, via the primary path 1. The earphone includes a cup-shaped housing 14 having a hole 15 that is covered by a sound-permeable cover, such as a grill, grid, or any other sound-permeable structure or material. The speaker 3 acts to radiate sound to the ear 12 and is disposed in the hole 15 of the housing 14 to form the earphone cavity 13 together. The cavity 13 is hermetically sealed or ventilated by any means such as an entrance / exit, a vent, or an opening. The microphone 4 is disposed in front of the speaker 3. The acoustic path 17 extends from the speaker 3 to the ear 12 and has a transfer characteristic approximated for noise suppression purposes by the transfer characteristic of the secondary path 2 extending from the speaker 3 to the microphone 4.

  In portable devices such as headphones, the space and energy that the ANC system can use is extremely limited. Analog circuits are often preferred in the design of ANC systems in portable devices because digital circuits consume too much space and energy. However, it is difficult to accurately model the secondary path with analog means alone, as analog circuits alone limit the potential complexity of the ANC system. Specifically, the analog filters used in ANC systems are either fixed filters or very simple adaptive filters because they are easy to assemble, consume less energy, and save space. There are many cases. Even in the system described above with reference to FIGS. 4, 5 and 7, when an analog circuit is used, it is hardly affected by the behavior of the secondary path (FIG. 4) or not (FIG. 5). And FIG. 7) showed good results. Further, the systems of FIGS. 5 and 7 are basically only slight variations, but the ANC filter transfer characteristics W (z), which together form the open loop transfer characteristics W (z) · S (z), A good estimate of the transfer characteristics required for the equivalent filter is possible based on the filter characteristics S (z) of the secondary path and on the acoustic characteristics of the headphones when attached to the listener's head And

  The ANC filter 5 normally has a transfer characteristic that tends to have a low gain at a low frequency such that the gain increases up to the maximum gain with respect to the frequency, and then the gain decreases with respect to the frequency and continues to decrease to reach the loop gain. . If the ANC filter 5 is high gain, the loop inherent to the ANC system, for example, keeps the system linear in the frequency range below 1 kHz, thus making any equivalent redundant in this frequency range. In the frequency range above 3 kHz, this ANC filter 5 has no boost or cut and therefore no linear effect. Since the ANC filter gain within this frequency range is substantially a loop gain, the transfer characteristic M (z) of the useful signal needs to be compensated by each filter, for example, a shelving filter added to the equivalent filter. Experience a boost at In the frequency range from 1 kHz to 3 kHz, both boost and cut may occur. For auditory impressions, boost is more disturbing than cut, so it is sufficient to compensate for the transfer characteristic boost with a similarly designed cut filter. If the gain of the ANC filter is greater than 3 kHz and 0 dB, there is no linearization effect, and therefore a second equivalent filter can be used in addition to the first equivalent filter and instead of the shelving filter.

As can be seen from the above discussion, at least two filters are used for compensation. FIG. 7 shows an exemplary ANC system that uses (at least) two filters 18 and 19 instead of a single filter 11 in the system of FIG. For example, in a treble cut shelving filter (for example, filter 18) having a filter transfer characteristic S ₁ (z) and a treble cut equivalent filter (for example, filter 19) having a transfer characteristic S ₂ (z), S (z) = S ₁ (z) · S ₂ (z). Alternatively, a treble boost equivalent filter can be implemented, for example, as the filter 18, and a treble cut equivalent filter can be implemented, for example, as the filter 19. When the useful signal transfer characteristic M (z) develops an even more complicated structure, for example, three filters having one treble cut shelving filter and two treble boost / cut equivalent filters can be used. The number of filters used is affected by many other situations, such as cost, filter noise behavior, headphone acoustic characteristics, system delay time, and space available to implement the system.

  FIG. 8 is a diagram schematically showing transfer characteristics a and b of a shelving filter applicable to the system of FIG. In particular, a first order treble boost (+9 dB) shelving filter (a) and a bass cut (-3 dB) shelving filter (b) are shown. FIG. 9 is a diagram schematically showing transfer characteristics c and d of an equivalent filter applicable to the system of FIG. One of the equivalent filters (c) provides a 9 dB boost at 1 kHz and the other filter (d) provides a 6 dB cutoff at 100 Hz with a higher Q, thus providing a sharper bandwidth.

  The range of the spectral shaping function is governed by the theory of linear filters, but the adjustment of these functions and the flexibility with which they are adjusted will vary depending on the topology of the circuit and the requirements that need to be fulfilled. A shelving filter is usually a simple first order filter that changes the relative gain between frequencies much higher and much lower than the corner frequency. The low or bass shelf is adjusted to affect the low frequency gain, but has little effect beyond its corner frequency. A high or treble shelf adjusts the gain at high frequencies only. On the other hand, a single equivalent filter implements a second order filter function. This has three adjustments: selection of the center frequency, adjustment of the quality (Q) factor that determines the sharpness of the bandwidth, and how much the selected center frequency is relative to the frequency (much) above and below the center frequency. Accompanied by a level or gain that determines whether it is boosted or cut.

FIG. 10 shows that the useful signal x [n] is transmitted through the filter 20 having the transfer characteristic S ₅ (z) or through the filter 21 having the transfer characteristic S ₆ (z). FIG. ₆ shows a combination of the systems shown in FIGS. 4 and 5, transmitted to both (here, for example, S (z) = S ₅ (z) · S ₆ (z)).

While various embodiments for practicing the invention have been disclosed, it will be apparent to those skilled in the art that various changes and modifications may be made to achieve the advantages of the invention without departing from the spirit and scope of the invention. Let's go. It will also be apparent to those skilled in the art that other components that perform similar functions may be appropriately substituted. Such variations on the inventive concept are covered by the appended claims.

Claims (20)

  A noise-removal audio playback system,
  A speaker connected to the speaker input path;
  A microphone acoustically coupled to the speaker via a secondary path and connected to a first end of a microphone output path, the secondary path having a secondary path transfer characteristic; ,
  A first subtractor coupled to a second end of the microphone output path and a first end of a first useful signal path;
  An active noise removal filter coupled downstream of the first subtractor;
  A second subtractor coupled between the active noise removal filter and the speaker input path, wherein the second subtractor is further coupled to a first end of a second useful signal path; With a second subtractor
Including
  The second end of the first useful signal path and the second end of the second useful signal path are interconnected and receive a useful signal reproduced by the speaker. The useful signal is a signal before subtraction in the first subtractor or the second subtractor,
  At least one of the first useful signal path or the second useful signal path includes one or more spectrum shaping filters, and at least one of the one or more spectrum shaping filters is the secondary path transmission. Has transfer characteristics modeling characteristics,
  A system, wherein a filter input signal is provided to the active denoising filter by the first subtractor, and the filter input signal approaches zero when denoising of the denoising filter is active.   The system of claim 1, wherein at least one of the one or more spectral shaping filters has a transfer characteristic that linearizes a microphone signal on the microphone output path with respect to the useful signal.   The system of claim 2, wherein the first useful signal path includes a first spectral shaping filter having a transfer characteristic that substantially matches the secondary path transfer characteristic.   The system of claim 3, wherein the first spectral shaping filter includes at least two sub-filters.   The system of claim 4, wherein at least one of the first spectral shaping filter or the at least two sub-filters of the first spectral shaping filter comprises an equivalent filter.   The system of claim 5, wherein the equivalent filter comprises a treble cut equivalent filter.   The system of claim 4, wherein the first spectral shaping filter or one of the at least two sub-filters of the first spectral shaping filter comprises a shelving filter.   The system of claim 7, wherein the shelving filter comprises a treble cut shelving filter.   At least one of the one or more spectrum shaping filters includes a first spectrum shaping filter, and the second useful signal path has a transfer characteristic that substantially matches the inverse of the secondary path transfer characteristic. A second spectral shaping filter that includes:
  The system of claim 2, wherein at least one of the active noise removal filter, the first spectral shaping filter, and the second spectral shaping filter includes an adaptive filter.   Each of the first useful signal path and the second useful signal path includes a respective spectral shaping filter having a respective transfer characteristic;
  The system of claim 1, wherein each of the first useful signal path and the second useful signal path is configured to be provided with the useful signal via their respective spectral shaping filters. .   A noise-removal audio playback method,
  Providing an input signal to a speaker, wherein the input signal is radiated acoustically by the speaker;
  Receiving a signal emitted by the speaker by a microphone, the microphone being acoustically coupled to the speaker via a secondary path, the microphone being configured to provide a microphone output signal; And that
  Subtracting the microphone output signal from the useful signal via a first subtractor to generate a filter input signal;
  Filtering the filter input signal with an active noise removal filter to generate an error signal;
  Subtracting the useful signal from the error signal via a second subtractor to generate a speaker input signal;
  Filtering the useful signal with one or more spectral shaping filters prior to subtraction of the useful signal from at least one of the microphone output signal or the error signal;
  Including
  Each of the first subtractor and the second subtractor is an analog device;
  The method wherein the filter input signal approaches zero when denoising of the denoising filter is active.   The method of claim 11, wherein the secondary path has a secondary path transfer characteristic, and the one or more spectral shaping filters model the secondary path transfer characteristic as a whole.   The method of claim 12, wherein the useful signal is filtered by a transfer characteristic that substantially matches the secondary path transfer characteristic prior to subtraction from the microphone output signal.   14. The method of claim 13, wherein the useful signal filtering comprises at least one of equivalent filtering or shelving filtering of the useful signal.   13. The method of claim 12, wherein the useful signal is filtered by a transfer characteristic that substantially matches the inverse of the secondary path transfer characteristic prior to subtraction from the error signal.   A noise-removal audio playback system,
  A first subtractor configured to receive an acoustic signal used to drive a speaker to generate an audible sound,
    The first subtractor is further configured to receive a microphone input signal, the microphone input signal including the audible sound received from the speaker and unwanted noise detected by the microphone in a listening space. Including
    The first subtractor is further configured to subtract the acoustic signal from the microphone input signal to generate a filter input signal; and
  An active noise removal filter in communication with the first subtractor, wherein the active noise removal filter is configured to generate an error signal based on the filter input signal; ,
  A second subtractor in communication with the active noise removal filter, wherein the second subtracter subtracts the acoustic signal from the error signal and outputs a speaker input signal to drive the speaker; A second subtractor configured to:
  At least one spectral shaping filter configured to receive and filter the acoustic signal, the at least one spectral shaping filter being one of the first subtractor or the second subtractor. At least one spectral shaping filter configured to output the filtered acoustic signal to at least one;
  Including
  Each of the first subtractor and the second subtractor is an analog device;
  The noise-removed sound reproduction system, wherein the filter input signal approaches zero when noise removal of the noise removal filter is active.   The at least one spectrum shaping filter includes a first spectrum shaping filter and a second spectrum shaping filter, and the first spectrum shaping filter outputs the filtered acoustic signal to the first subtractor. The second spectrum shaping filter is configured to output the filtered acoustic signal to the second subtractor, and the filtering by the first spectrum shaping filter and the The noise-reduced sound reproduction system according to claim 16, which is different from filtering by the second spectrum shaping filter.   The first spectral shaping filter includes a filter coefficient that represents the inverse of the estimated secondary path transfer characteristic between the speaker and the listening position, and the second spectral shaping filter includes the estimated secondary The noise-removed voice reproduction system according to claim 17, comprising a filter coefficient representing a path transfer characteristic.   The at least one spectrum shaping filter includes a first shelving filter configured as a boost filter operable at a first center frequency and a second filter configured as a cut filter operable at a second center frequency. The noise-removed sound reproduction system according to claim 16, further comprising a shelving filter, wherein the first center frequency is higher than the second center frequency.   The noise-removed sound reproduction system according to claim 16, wherein each of the at least one spectrum shaping filter and the active noise removal filter is an analog device.

Images