Sound emitting device

Abstract

A sound emitting device that is worn by a user, and has a first driver unit 41 and a second driver unit 42, wherein the sound emitting device is of a configuration in which noise arriving from the exterior can arrive at the user's ears, sound emitted from the first driver unit is controlled so as to be a first acoustic signal that is based on a first signal obtained by applying a first bandpass filter to an acoustic signal that is based on the noise, sound emitted from the second driver unit is controlled so as to be a second acoustic signal that is based on a second signal obtained by applying a second bandpass filter to the acoustic signal that is based on the noise, and a diaphragm of the first driver unit is larger than a diaphragm of the second driver unit.

Description

sound emitting device

[0001] The disclosed technology relates to noise cancellation technology.

[0002] In the prior art, closed-back headphones were the mainstream for headphones requiring noise cancellation. For information on closed-back headphones, please refer to Non-Patent Document 1. Closed-back headphones can reduce noise at a predetermined frequency through their physical structure. Therefore, in the prior art, noise cancellation only needed to consider frequency bands other than that predetermined frequency.

[0003] Audio-Technica Corporation, "Understanding Headphones and Earphones," [online], [Accessed October 18, 2024], Internet <URL: https: / / www.audio-technica.co.jp / headphone / navi / whatis / 02-03.php>

[0004] However, in the case of headphones, smart glasses, and earphones that have a structure that cannot reduce a specific frequency through physical means, noise cancellation covering a wide frequency range is necessary.

[0005] The disclosed technology aims to provide a noise-canceling device, method, and program that can perform noise cancellation over a wider frequency band than conventional methods.

[0006] One aspect of the disclosed technology is a user-worn sound-emitting device comprising a first driver unit and a second driver unit, wherein the sound-emitting device is configured such that external noise can reach the user's ears, the sound emitted from the first driver unit is controlled to be a first acoustic signal based on a first signal obtained by applying a first bandpass filter to an acoustic signal based on noise, and the sound emitted from the second driver unit is controlled to be a second acoustic signal based on a second signal obtained by applying a second bandpass filter to an acoustic signal based on noise.

[0007] Furthermore, the diaphragm of the first driver unit is larger than the diaphragm of the second driver unit.

[0008] According to the disclosed technology, noise cancellation can be performed over a wider frequency band than before.

[0009] Figure 1 shows an example of the functional configuration of the sound emission device according to the first embodiment. Figure 2 shows an example of the processing procedure for the sound emission method. Figures 3(a) to 3(d) each show an example of a filter. Figure 4 shows an example of the functional configuration of the sound emission device according to the second embodiment. Figure 5 shows an example of the functional configuration of the sound emission device according to the third embodiment. Figure 6 shows an example of the functional configuration of the sound emission device according to the fourth embodiment. Figure 7 shows an example of the functional configuration of the sound emission device according to the fifth embodiment. Figure 8 shows an example of the functional configuration of the sound emission device according to the fifth embodiment. Figure 9 shows an example of the functional configuration of the sound emission device according to the sixth embodiment. Figures 10(a) to 10(d) each show an example of a filter. Figure 11 shows an example of the functional configuration of the sound emission device according to the seventh embodiment. Figures 12(a) and 12(b) each show an example of a filter. Figure 13 is a diagram for explaining an experimental example. Figure 14 is a diagram for explaining an experimental example. Figure 15 is a diagram for explaining the effect of the sound emission device. Figure 16 shows an example of the functional configuration of a computer. Figure 17 is a diagram illustrating an additional embodiment. Figure 18 is a diagram illustrating an additional embodiment. Figure 19 is a diagram illustrating an example of the position of the diaphragm of the first driver unit and the diaphragm of the second driver unit. Figure 20 is a diagram illustrating an example of the arrangement of the first driver unit and the second driver unit. Figure 21 is a diagram illustrating an experimental example of the additional embodiment. Figure 22 is a diagram illustrating an experimental example of the additional embodiment. Figure 23 is a diagram illustrating an example of the arrangement of the first driver unit and the second driver unit. Figure 24 is a diagram illustrating an example of the arrangement of the first driver unit and the second driver unit. Figure 25 is a diagram illustrating an experimental example of the additional embodiment. Figure 26 is a diagram illustrating an experimental example of the additional embodiment. Figure 27 is a diagram illustrating an additional embodiment. Figure 28 is a diagram illustrating an additional embodiment.

[0010] Hereinafter, embodiments of the disclosed technology will be described with reference to the drawings. In the drawings, components having the same function are denoted by the same reference numerals, and redundant description will be omitted.

[0011] [First Embodiment] The sound-emitting device is a device with little sound insulation by a housing such as an open-ear type earphone, a shoulder speaker, an audio glass, or smart glasses, and a device that employs a plurality of types of driver units.

[0012] The open-ear type earphone may be an open-ear type earphone of any shape.

[0013] For example, the open-ear type earphone may be a clip type that夹入 is夹入 the lower part of the ear. As such a clip type open-ear type earphone, for example, the one described in Reference 1 is known.

[0014] [Reference 1] HUAWEI, "HUAWEI FreeClip", [online], [searched on November 5, 2024], Internet <URL: https: / / consumer.huawei.com / jp / audio / freeclip / >. Also, the open-ear type earphone may have a shape including a sound-emitting part arranged in the auricle direction from the upper part of the ear, a structural part arranged in the back direction from the upper part of the ear, and a connecting part between the two structural parts. As such an open-ear type earphone, for example, the one described in Reference 2 is known.

[0015] [Reference 2] Anker Japan Co., Ltd., "Soundcore AeroFit", [online], [searched on November 5, 2024], Internet <URL: https: / / www.ankerjapan.com / products / a3872?variant=4335484<5970593>. The open-ear type earphone may be a local sound-emitting type. The local sound-emitting type may be realized by direction control or by using sound waves of an inverse phase that are emitted to the side opposite to the ear direction.

[0016] It should be noted that the phrase "夹入" in the original text seems to be incorrect or incomplete. I've translated it as "夹入" as it is in the original, but it might need to be corrected in the source text for a more accurate translation.As shown in Figure 1, the sound emission device of the first embodiment includes, for example, a reference microphone RM, a first filter processing unit 11, a second filter processing unit 12, a third filter processing unit 21, a fourth filter processing unit 22, a noise cancellation unit 3, a first driver unit 41, a second driver unit 42, and an error microphone EM. Alternatively, the sound emission device may include only the noise cancellation unit 3, the first driver unit 41, and the second driver unit 42.

[0017] The sound emission method is achieved, for example, by having each component of the sound emission device perform the processing shown in steps S11 to S42 in Figure 2.

[0018] Note that the symbols used in the text are - The symbols " and " and " should ideally be placed directly above the character immediately following them, but due to limitations in text notation, they are placed directly before the character. In mathematical formulas, these symbols are placed in their original position, that is, directly above the character. For example, in the sentence " - "X" is written in mathematical formulas as follows: The following describes each component of the sound emission device.

[0019] [[Reference Microphone RM]] In Figure 1, the reference microphone RM is denoted as "RM".

[0020] The noise emitted from the noise source reaches the reference microphone RM. In Figure 1, the signal corresponding to the noise emitted from the noise source is denoted as x(n).

[0021] The reference microphone RM converts the incoming noise into a signal. The converted signal is denoted as r(n).

[0022] The signal x(n) corresponding to the noise is affected by the characteristic R(z) from the noise source to the AD conversion of the reference microphone RM. "R(z)" in Figure 1 illustrates this.

[0023] The signal r(n) obtained from the reference microphone is output to the first filter processing unit 11, the second filter processing unit 12, and the noise cancellation unit 3. Since the signal r(n) contains noise, it can also be called a noise-based acoustic signal.

[0024] [[First Filter Processing Unit 11 and Second Filter Processing Unit 12]] r(n) is input to the first filter processing unit 11.

[0025] First, the first filter processing unit 11 performs a process on r(n) using the first band-pass filter H wf (z) (step S11). Then, the first filter processing unit 11 performs a process on the signal after that process using a filter - S wf (z) that simulates the characteristics of a so-called secondary path.

[0026] The signal ~r wf (n) after that process is output to the noise canceling unit 3. ​​​​​​​​​​​​​​​​​​​​​​First, the second filter processing unit 12 applies a second bandpass filter H to r(n). tw (z) is used for processing (step S12). Then, the second filter processing unit 12 applies a filter to the processed signal that simulates the characteristics of a so-called secondary path. - S tw Perform processing using (z).

[0031] The signal after processing ~r tw (n) is output to the noise cancellation unit 3.

[0032] A filter that simulates the characteristics of a secondary path. - S tw (z) is, more specifically, the characteristic S from the DA conversion of the second driver unit 42 to the AD conversion of the microphone EM. tw This is a filter that simulates (z). For example, this characteristic is measured in advance using a dummy head or the like.

[0033] Second bandpass filter H tw (z) is predetermined considering the characteristics and position of the second driver unit 42. Second bandpass filter H tw (z) may also be predetermined by considering the characteristics and position of the first driver unit 41. Specifically, the F0 (lowest frequency band), frequency characteristics (amplitude, phase, and group delay), harmonic distortion, efficiency, power handling capacity, distance and frequency characteristics (amplitude, phase, and group delay) of the second driver unit 41 and the second driver unit 42 may be considered. For example, the second bandpass filter H tw (z) may be determined by taking into account the preferred frequency range determined according to the size of the driver unit, and the distance between the ear and the area where it can be placed, which is determined by the size.

[0034] First bandpass filter H wf (z) and the second bandpass filter H tw(z) may be a filter used in a typical two-way crossover circuit, as illustrated in Figure 3(a). An example of a filter used in a typical two-way crossover circuit is the Linkwitz-Riley filter.

[0035] Also, the first bandpass filter H wf (z) and the second bandpass filter H tw (z) may be a high-pass filter, as illustrated in Figure 3(b). Note that if the second driver unit 42 is a smaller driver unit than the first driver unit 41, the second band-pass filter H tw The cutoff frequency of (z) is set by the first bandpass filter H wf The frequency may be higher than the cutoff frequency of (z).

[0036] Also, the first bandpass filter H wf (z) and the second bandpass filter H tw (z) may be a weighted bandpass, as illustrated in Figure 3(c).

[0037] Furthermore, as illustrated in Figure 3(d), the first bandpass filter H wf (z) and the second bandpass filter H tw One of (z) may be a subband filter and the other a pass-through filter.

[0038] [[Third filter processing unit 21 and fourth filter processing unit 22]] The error signal e(n) is input to the third filter processing unit 21.

[0039] The third filter processing unit 21 applies the first bandpass filter H to the error signal e(n). wf (z) is used for processing (step S21). The first bandpass filter H wf (z) is the first bandpass filter H used in the first filtering unit 11. wf This is the same filter as (z).

[0040] The processed signal is output to the noise cancellation unit 3.

[0041] The error signal e(n) is input to the fourth filter processing unit 22.

[0042] The fourth filter processing unit 22 applies a second bandpass filter H to the error signal e(n). tw Processing is performed using (z) (step S22). Second bandpass filter H tw (z) is the second bandpass filter H used in the second filtering unit 12. tw This is the same filter as (z).

[0043] The processed signal is output to the noise cancellation unit 3.

[0044] [[Noise Cancellation Unit 3]] The noise cancellation unit 3 receives r(n) and the signal after filtering by the first filter processing unit 11 ~r wf (n) and the signal after filtering by the second filter processing unit 12 ~r tw (n), the signal after filtering by the third filter processing unit 21, and the signal after filtering by the fourth filter processing unit 22 are input.

[0045] First, the noise cancellation unit 3 generates an adaptively updated filter. Adaptive algorithms such as LMS, NLMS, MLMS, and Delayless SubbandLMS are used to estimate the filter. Of course, other appropriate adaptive algorithms may also be used.

[0046] Specifically, the noise cancellation unit 3 processes the signal after filtering by the first filter processing unit 11 ~r wf A filter W generates a secondary sound source signal for active noise control using (n) and the signal after filtering by the third filter processing unit 21. wf Estimate (z).

[0047] Thus, filterW wf(z) is generated from the first signal obtained by applying the first bandpass filter to the noise-based acoustic signal and the signal obtained by applying the first bandpass filter to the error signal. The noise-based acoustic signal is r(n). In this example, the first signal is the signal ~r after filtering by the first filter processing unit 11. wf (n) is the case, and the signal obtained by applying the first bandpass filter to the error signal is the signal after filtering by the third filter processing unit 21.

[0048] Furthermore, the noise cancellation unit 3 processes the signal after filtering by the second filter processing unit 12 ~r tw A filter W generates a secondary sound source signal for active noise control using (n) and the signal after filtering by the fourth filter processing unit 22. tf Estimate (z).

[0049] Thus, filterW tf (z) is generated from a second signal obtained by applying a second bandpass filter to the noise-based acoustic signal and a signal obtained by applying a second bandpass filter to the error signal. In this example, the second signal is the signal ~r after filtering by the second filter processing unit 12. tw (n) is the case, and the signal obtained by applying a second bandpass filter to the error signal is the signal after filtering by the fourth filter processing unit 22.

[0050] Then, the noise cancellation unit 3 applies the filter W to r(n). wf The first acoustic signal y is processed using (z). wf (n) is generated. The first acoustic signal y wf (n) is output to the first driver unit 41.

[0051] Furthermore, the noise cancellation unit 3 applies the filter W to r(n). tw The second acoustic signal y is processed using (z). tw (n) is generated. Second acoustic signal y tw (n) is output to the second driver unit 42.

[0052] In this way, the noise cancellation unit 3 controls the first driver unit to emit a first acoustic signal based on a first signal obtained by applying a first bandpass filter to the noise-based acoustic signal (step S3). The noise cancellation unit 3 also controls the second driver unit to emit a second acoustic signal based on a second signal obtained by applying a second bandpass filter to the noise-based acoustic signal (step S3).

[0053] [[First driver unit 41 and second driver unit 42]] In Figure 1, the first driver unit 41 is denoted as "DU41".

[0054] The first driver unit 41 receives the first acoustic signal y wf (n) is entered.

[0055] The first driver unit 41 has the first characteristic.

[0056] The first driver unit 41 receives the first acoustic signal y wf (n) emits a sound corresponding to (n) (step S41).

[0057] In Figure 1, the second driver unit 42 is labeled "DU42".

[0058] The second driver unit 42 receives the second acoustic signal y tw (n) is entered.

[0059] The second driver unit 42 has a second characteristic.

[0060] The second driver unit 42 receives the second acoustic signal y tw The sound corresponding to (n) is emitted (step S42).

[0061] The first driver unit 41 and the second driver unit 42 may be conventional speakers that transmit sound by vibrating the air, bone conduction speakers, or cartilage conduction speakers. Conventional speakers that transmit sound by vibrating the air may be dynamic speakers, BA speakers, or electrostatic speakers.

[0062] As will be described later, the first driver unit 41 and the second driver unit 42 may be driver units of different sizes.

[0063] For example, the first driver unit 41 and the second driver unit 42 may be two driver units with different characteristics (e.g., a woofer and a tweeter) included in a 2-way coaxial speaker. The first driver unit 41 and the second driver unit 42 may be a headrest speaker and a headphone earphone, respectively.

[0064] For example, the second driver unit 42 may be a smaller driver unit than the first driver unit.

[0065] [[Error Microphone EM]] In Figure 1, the error microphone EM is denoted as "EM".

[0066] The sound emitted by the first driver unit 41 and the second driver unit 42 combines with the noise signal d(n) in the actual acoustic space and reaches the error microphone EM.

[0067] The error microphone EM converts the incoming sound into a signal. The converted signal is denoted as the error signal e(n). The error signal e(n) is output to the third filter processing unit 21 and the fourth filter processing unit 22.

[0068] The noise emitted from the noise source is affected by the characteristic P(z) from the noise source to the AD conversion of the error microphone EM. "P(z)" in Figure 1 illustrates this.

[0069] The sound emitted by the first driver unit 41 has characteristics S from the DA conversion of the first driver unit 41 to the AD conversion of the microphone EM. wf (z) is affected. wf (z) indicates this.

[0070] The sound emitted by the second driver unit 42 has characteristics S from the DA conversion of the second driver unit 42 to the AD conversion of the microphone EM. tw (z) is affected. tw (z) indicates this.

[0071] This type of sound emission device and method enables noise cancellation over a wider frequency range than conventional methods.

[0072] Conventional earphones and headphones that block the ear canal physically block high-frequency noise above 1kHz. Therefore, ANC (Active Noise Cancellation) only needed to suppress low-frequency noise below 1kHz. However, open-ear earphones and headphones that do not block the ear canal are designed to allow some external sounds to be heard, so high-frequency noise above 1kHz also reaches the ear.

[0073] Incidentally, the human ear is particularly sensitive to sounds around 3kHz, and this frequency range is the most audible. In order to effectively suppress noise with open-ear headphones, it was also necessary to suppress high-frequency bands above 1kHz using ANC.

[0074] Incidentally, larger driver units can cancel out lower frequencies better than smaller driver units. Also, larger driver units can produce higher sound pressure without distortion than smaller driver units. However, the delay in the secondary path increases at high frequencies. Specifically, if you try to install a large driver unit so that it does not block the ear, the distance between the driver unit and the error microphone or the ear / ear canal entrance becomes greater. In addition, high-frequency components require the diaphragm to move at high speeds, and the group delay increases at high frequencies due to the inertia caused by the large mass. Therefore, it is disadvantageous for high-frequency sound cancellation.

[0075] On the other hand, small driver units can be positioned near the error microphone or the ear or ear canal entrance without blocking the ear, thus reducing secondary path delay. However, small driver units have a higher F0 compared to large driver units, and group delay increases around F0, making it difficult to cancel low frequencies. If the error signal contains low frequencies, distortion may occur as the driver attempts to forcibly produce a low-frequency cancellation signal. For this reason, small driver units have difficulty producing high sound pressure.

[0076] Therefore, the first driver unit 41 and the second driver unit 42 are made of driver units of different sizes. The larger driver unit is used to silence low frequencies, and the smaller driver unit is used to silence high frequencies. This makes effective noise cancellation possible.

[0077] Furthermore, when the sound-emitting device is attached, the smaller driver unit may be positioned closer to the ear than the larger driver unit. This allows for more effective sound cancellation.

[0078] Furthermore, in order to cancel out high frequencies, the sound waves used for noise cancellation must reach the error microphone as quickly as possible in relation to the noise. In other words, the processing delay (the sum of the delay Dr of the reference path and the delay Ds of the secondary path) must be kept small compared to the noise propagation delay (the delay of the primary path).

[0079] As illustrated in Figure 15, a tweeter, which is an example of a small driver unit, has a short secondary path and low group delay above 1 kHz. For this reason, it is preferable to have the tweeter cancel out high-frequency noise above 1 kHz. On the other hand, a woofer, which is an example of a large driver unit, has a large diaphragm and low group delay at low frequencies. For this reason, it is preferable to have the woofer cancel out low-frequency noise below 1 kHz.

[0080] Processing delays can be divided into acoustic delays and mechanical delays.

[0081] By using driver units of different sizes for the first driver unit 41 and the second driver unit 42, and having the larger driver unit silence low frequencies and the smaller driver unit silence high frequencies, the mechanical delay in the secondary path can be improved. This is because reducing the size of the driver unit reduces the mechanical delay in the secondary path. Thus, the mechanical delay can vary depending on the size of the driver unit, in other words, the size, structure and material of its diaphragm, and the structure and material of the related driver unit (e.g., edge and damper).

[0082] If the second driver unit 42 is smaller than the first driver unit 41, the acoustic delay of the secondary path related to the second driver unit 42 can be reduced by bringing the distance between the error microphone EM and the second driver unit 42 closer. The second driver unit performs high-frequency noise suppression by the second filter processing unit 12 and the fourth filter processing unit 22, so the high-frequency noise suppression performance is improved by reducing the high-frequency group delay.

[0083] [Second Embodiment] The sound emission device and method of the second embodiment will be described below. The explanation will focus on the parts that differ from the first embodiment. Parts that are the same as those of the first embodiment will be omitted.

[0084] As shown in Figure 4, the first filter processing unit 11 and the third filter processing unit 21 are the first bandpass filter H wf To perform processing using (z), two bandpass filters H wf,1 (z), H wf,2 You may also perform processing using (z).

[0085] Similarly, the second filter processing unit 12 and the fourth filter processing unit 22 control the second bandpass filter H tw To perform processing using (z), two bandpass filters H tw,1 (z), H tw,2 You may also perform processing using (z).

[0086] Thus, multiple frequency band limiting filters may be cascaded. This allows for more precise filtering.

[0087] [Third Embodiment] The sound emission device and method of the third embodiment will be described below. The explanation will focus on the parts that differ from the first embodiment. Parts that are the same as those of the first embodiment will not be explained again.

[0088] As shown in Figure 5, the first filter processing unit 11 first simulates the characteristics of the secondary path using a filter. - S wf The signal is processed using (z), and then two bandpass filters H are applied to the resulting signal. wf,1 (z), Hwf,2 Perform each of the processes using (z) to generate two post - processed signals ~r wf,1 (n), ~r wf,2 (n). For example, H wf,1 (z) is a band - pass filter that passes a band from 100 - 500 Hz, and H wf,2 (z) is a band - pass filter that passes a band from 500 - 900 Hz.

[0089] Also, the third filter processing unit 21 may perform each of the processes using two band - pass filters H wf,1 (z), H wf,2 (z) on the error signal e(n) to generate two post - processed signals.

[0090] In this case, the noise canceling unit 3 generates a filter based on the FxLMS algorithm using the post - processed signal ~r wf,1 (n) processed by the first filter processing unit 11 using H wf,1 (z) and the signal processed by the third filter processing unit 21 using H wf,1 (z). Also, the noise canceling unit 3 generates a filter based on the FxLMS algorithm using the post - processed signal ~r wf,2 (n) processed by the first filter processing unit 11 using H wf,2 (z) and the signal processed by the third filter processing unit 21 using H wf,2 (z). The noise canceling unit 3 finally calculates the filter W wf (z) using these generated filters.

[0091] Similarly, the second filter processing unit 12 first performs a process using a filter - S tw (z) that simulates the characteristics of the secondary path, and then performs each of the processes using two band - pass filters H tw,1 (z), H tw,2 (z) on the signal after that process to generate two post - processed signals ~r tw,1 (n), ~r tw,2 (n). For example, H tw,1(z) is a band-pass filter that passes a band of 900 - 1300 Hz, and H tw,2 (z) is a band-pass filter that passes a band of 1300 - 1800 Hz.

[0092] Also, the fourth filter processing unit 22 may perform each of the processes using two band-pass filters H tw,1 (z), H tw,2 (z) and generate two processed signals.

[0093] In this case, the noise canceling unit 3 generates a filter based on the FxLMS algorithm using the processed signal ~r tw,1 (n) processed using H tw,1 (z) by the second filter processing unit 12 and the signal processed using H tw,1 (z) by the fourth filter processing unit 22. Also, the noise canceling unit 3 generates a filter based on the FxLMS algorithm using the processed signal ~r tw,2 (n) processed using H tw,2 (z) by the second filter processing unit 12 and the signal processed using H tw,2 (z) by the fourth filter processing unit 22. The noise canceling unit 3 finally calculates the filter W tw (z) using these generated filters.

[0094] The noise canceling unit 3 performs a process using the filter W wf (z) on x(n) and generates the first acoustic signal y wf (n). The first acoustic signal y wf (n) is output to the first driver unit 41.

[0095] The noise canceling unit 3 performs a process using the filter W tw (z) on x(n) and generates the second acoustic signal y tw (n). The second acoustic signal y tw (n) is output to the second driver unit 42.

[0096] In this way, by connecting multiple filters in parallel, it becomes possible to estimate the noise control filter for each frequency band.

[0097] [Fourth Embodiment] The sound emission device and method of the fourth embodiment will be described below. The explanation will focus on the parts that differ from the first and second embodiments. Parts that are the same as those of the first and second embodiments will not be explained again.

[0098] As shown in Figure 6, a part of the bandpass filter is the noise control filter W wf (z), W tw (z) may be placed before or after it. In the example in Figure 6, the noise control filter W tw After (z), a bandpass filter H tw,2 (z) is positioned. Also, a bandpass filter H tw,1 After (z), a bandpass filter - H tw,2 (z) is placed. Bandpass filter - H tw,2 (z) is a bandpass filter H tw,2 This is a filter that simulates the characteristics of (z). It is a bandpass filter. - H tw,2 (z) is a bandpass filter H tw,2 It may be the same filter as (z).

[0099] In this case, the second filter processing unit 12 is a bandpass filter H tw,1 After processing using (z), a bandpass filter is applied. - H tw,2 Perform processing using (z).

[0100] The fifth filter processing unit 52 processes the second acoustic signal y generated by the noise cancellation unit 3. tw For (n), bandpass filter H tw,2 Processing is performed using (z). The signal after this processing is output to the second driver unit 42.

[0101] Note that the bandpass filter H tw,2 (z) may be an analog filter instead of a digital filter. Also, a bandpass filter Htw,2 (z) may be realized by acoustic phenomena based on the enclosure design, such as Hemholtz resonance, or by sound insulation. In this case, the sound emission device does not need to include the fifth filter processing unit 52 shown by the solid line in Figure 6. In this case, the fifth filter processing unit 52 can be considered to be virtually located at the position shown by the dashed line in Figure 6.

[0102] Note that characteristic S in this case tw (z) is a bandpass filter H realized by acoustic phenomena based on the design of the enclosure, such as an analog filter or Hemholtz resonance, or by sound insulation, based on the characteristics from the DA conversion of the second driver unit 42 to the AD conversion of the error microphone EM. tw,2 This represents the properties after removing the influence of the properties of (z).

[0103] In this way, a part of the bandpass filter is used as the noise control filter W wf (z), W tw By placing it before and after (z), the actual silencing signal can be directly altered.

[0104] [Fifth Embodiment] The sound emission device and method of the fifth embodiment will be described below. The description will focus on the parts that differ from the first, second, and fourth embodiments. Parts that are the same as those of the first, second, and fourth embodiments will be omitted.

[0105] As shown in Figures 7 and 8, there may be only one noise control filter. Reducing the number of noise control filters can reduce the computational load.

[0106] In these examples, the first filter processing unit 11 is a bandpass filter H wf,1 After processing using (z), a bandpass filter is applied. - H wf,2 Processing is performed using (z). The first filter processing unit 11 applies a filter to the processed signal that simulates the characteristics of a so-called secondary path. - S wf The process is performed using (z). The resulting signal ~r wf (n) is output to the noise cancellation unit 3.

[0107] Furthermore, the second filtering unit 12 is a bandpass filter H tw,1 After processing using (z), a bandpass filter is applied. - H tw,2 Processing is performed using (z). The second filter processing unit 12 applies a filter to the processed signal that simulates the characteristics of a so-called secondary path. - S tw The process is performed using (z). The resulting signal ~r tw (n) is output to the noise cancellation unit 3.

[0108] In the example shown in Figure 7, the noise cancellation unit 3 processes the signal after filtering by the first filter processing unit 11 ~r wf A filter W generates a secondary sound source signal for active noise control using (n) and the signal after filtering by the third filter processing unit 21. wf Estimate (z).

[0109] Furthermore, the noise cancellation unit 3 processes the signal after filtering by the second filter processing unit 12 ~r tw A filter W generates a secondary sound source signal for active noise control using (n) and the signal after filtering by the fourth filter processing unit 22. tf Estimate (z).

[0110] The noise cancellation section 3 is a filter W wf (z) and filter W tf It is combined with (z) to generate a single noise control filter W(z).

[0111] The noise cancellation unit 3 may generate a noise control filter W(z) based on, for example, the following equation.

[0112] W(f)=α(f)W wf (f) + (1-α(f))W tf (f) α(f)∈[0.0, 1.0] is the mixing coefficient at frequency f. α(f) may be manually set in advance to obtain the desired result.

[0113] In the example shown in Figure 8, the noise cancellation unit 3 processes the signal after filtering by the first filter processing unit 11 ~rwf (n) and the signal after filtering by the second filter processing unit 12 ~r tw (n) is combined with (n). The noise cancellation unit 3 also combines the signal after filtering by the third filter processing unit 21 with the signal after filtering by the fourth filter processing unit 22. The noise cancellation unit 3 uses these combined signals to generate a single noise control filter W(z).

[0114] Finally, the noise cancellation unit 3 processes x(n) using the filter W(z), and the acoustic signal y wf (n), y tw Generate (n).

[0115] The fifth filter processing unit 52 processes the second acoustic signal y generated by the noise cancellation unit 3. tw For (n), bandpass filter H tw,2 Processing is performed using (z). The signal after this processing is output to the second driver unit 42.

[0116] The sixth filter processing unit 51 processes the second acoustic signal y generated by the noise cancellation unit 3. wf For (n), bandpass filter H wf,2 Processing is performed using (z). The signal after this processing is output to the first driver unit 41.

[0117] Note that the bandpass filter H wf,2 (z) and - H wf,2 (z) is optional. In other words, the sound emission device does not need to have the sixth filter processing unit 51. Also, the first filter processing unit 11 is a bandpass filter. - H wf,2 It is not necessary to perform the processing using (z).

[0118] Note that the bandpass filter H wf,2 (z), H tw,2 (z) may be an analog filter instead of a digital filter. Also, a bandpass filter H wf,2 (z), H tw,2(z) may be realized by acoustic phenomena based on the enclosure design, such as Hemholtz resonance, or by sound insulation. In this case, the sound emission device does not need to include the fifth filter processing unit 52 and the sixth filter processing unit 51 shown by solid lines in Figures 7 and 8. In this case, the fifth filter processing unit 52 and the sixth filter processing unit 51 can be considered to be virtually located at the positions shown by dashed lines in Figures 7 and 8.

[0119] [Sixth Embodiment] The sound emission device and method of the sixth embodiment will be described below. The explanation will focus on the parts that differ from the first to fifth embodiments. Parts that are the same as those of the first to fifth embodiments will be omitted.

[0120] In the first to fifth embodiments, the number of error microphones may be two or more. Increasing the number of error microphones can widen the sound-canceling frequency range.

[0121] Furthermore, in the first to fifth embodiments, one of the first filter processing unit 11 and the second filter processing unit 12 does not have to perform bandpass filtering. Also, one of the third filter processing unit 21 and the fourth filter processing unit 22 does not have to perform bandpass filtering. Note that if bandpass filtering is not performed in the first filter processing unit 11, then bandpass filtering is not performed in the third filter processing unit 21. Also, if bandpass filtering is not performed in the second filter processing unit 12, then bandpass filtering is not performed in the fourth filter processing unit 22.

[0122] Furthermore, the bandpass filtering in the third filter processing unit 21 and the fourth filter processing unit 22 may be realized by acoustic phenomena based on the enclosure design, such as Hemholtz resonance, or by sound insulation. In this case, the sound emission device does not need to include the third filter processing unit 21 and the fourth filter processing unit 22.

[0123] In the example shown in Figure 9, two error microphones, EM1 and EM2, are used. In this example, the signal picked up by error microphone EM1 is predominantly low-frequency. Error microphone EM1 is covered with a material that provides sound insulation down to relatively low frequencies. Alternatively, error microphone EM1 is constructed to provide sound insulation down to relatively low frequencies. This characteristic of sound insulation down to relatively low frequencies is denoted as H1(z).

[0124] In addition, in the example shown in Figure 9, the first filter processing unit 11 is a bandpass filter. - The H1(z) signal is being processed. However, the second filter processing unit 12 does not perform bandpass filtering.

[0125] In Figure 9, the "filter estimation unit" is denoted as "FEP". In Figure 9, P1(z) shows the characteristics from the noise source to the AD conversion of the error microphone EM1. P2(z) shows the characteristics from the noise source to the AD conversion of the error microphone EM2. wf (z) shows the characteristics from the DA conversion of the first driver unit 41 to the AD conversion of the error microphone EM1. wf (z) shows the characteristics from the DA conversion of the first driver unit 41 to the AD conversion of the error microphone EM2. tw (z) shows the characteristics from the DA conversion of the second driver unit 42 to the AD conversion of the error microphone EM2. tw (z) shows the characteristics from the DA conversion of the second driver unit 42 to the AD conversion of the error microphone EM1. - C wf (z) is C wf This filter simulates the characteristics of (z). - C tw (z) is C tw This filter simulates the characteristics of (z).

[0126] [Seventh Embodiment] The sound emission device and method of the seventh embodiment will be described below. The explanation will focus on the parts that differ from the first to sixth embodiments. Parts that are the same as those of the first to sixth embodiments will be omitted.

[0127] In the sound emission device and method of the first to sixth embodiments, the filter determination unit 6, shown by the dashed line in Figures 1 and 4 to 9, may determine a bandpass filter (for example, at least one of the bandpass filters used in the first to sixth filter processing units 11 to 51) based on an external sound signal. In this case, the first to sixth filter processing units 11 to 51 perform processing using the bandpass filter determined by the filter determination unit 6.

[0128] For example, the filter determination unit 6 may change parameters such as the coefficients and cutoff frequency of the bandpass filter based on an external sound signal.

[0129] Furthermore, the filter determination unit 6 may perform the processing described in (Example 1) to (Example 3).

[0130] (Example 1) A set of bandpass filters with a low cutoff frequency (H wf 1 (z), H tw 1 (z)) and a pair of bandpass filters with a high cutoff frequency (H wf 2 (z), H tw 2 (z)) is predetermined.

[0131] As shown in Figure 10(a), the bandpass filter H wf 1 (z) is the cutoff frequency f c1 This is a 700Hz low-pass filter, and a band-pass filter H tw 1 (z) is the cutoff frequency f c1 This is a 700Hz high-pass filter. As shown in Figure 10(b), the band-pass filter H wf 2 (z) is the cutoff frequency f c2 This is a 1000Hz low-pass filter, and a band-pass filter H tw 2 (z) is the cutoff frequency f c2 This is a 1000Hz high-pass filter.

[0132] The filter determination unit 6 receives an external audio signal. Examples of external audio signals are signals x(n), r(n), and e(n). Signals obtained from other microphones may also be used as the external audio signal.

[0133] The filter determination unit 6 calculates the value of a predetermined feature from the input external sound signal. Examples of predetermined feature quantities include sound pressure level, A-weighted sound pressure level, and at least one of the spectral centroid. In the following explanation, the case where the predetermined feature quantity is the spectral centroid will be used as an example.

[0134] The filter determination unit 6 determines whether the calculated value is smaller than a predetermined threshold.

[0135] If the calculated value is smaller than a predetermined threshold, it can be determined that low frequencies are dominant, and the filter determination unit 6 determines a set of bandpass filters with a low cutoff frequency (H wf 1 (z), H tw 1 The filter determination unit 6 decides to use (z). Otherwise, it decides to use a set of bandpass filters with a high cutoff frequency (H wf 2 (z), H tw 2 We decide to use (z)).

[0136] The first filter processing unit 11 and the third filter processing unit 21 are the first bandpass filter H wf (z) is the bandpass filter H determined by the filter determination unit 6. wf 1 Perform processing using (z).

[0137] The second filter processing unit 12 and the fourth filter processing unit 22 are second bandpass filters H tw (z) is the bandpass filter H determined by the filter determination unit 6. wf 2 Perform processing using (z).

[0138] In the case of noise dominated by low frequencies, such as noise inside trains, airplanes, and cars, a set of bandpass filters with a low cutoff frequency (H wf1 (z), H tw 1 It is preferable to use (z). Also, in the case of noise where high frequencies are dominant, such as noise in urban areas or noise inside stores, a set of bandpass filters with a high cutoff frequency (H wf 2 (z), H tw 2 It is preferable to use (z)).

[0139] The above process allows for the use of an appropriate bandpass filter based on the external audio signal.

[0140] (Example 2) Similar to Example 1, a set of bandpass filters with a low cutoff frequency (H wf 1 (z), H tw 1 (z)) and a pair of bandpass filters with a high cutoff frequency (H wf 2 (z), H tw 2 (z)) is predetermined.

[0141] The filter determination unit 6 receives an external audio signal. Examples of external audio signals are signals x(n), r(n), and e(n). Signals obtained from other microphones may also be used as the external audio signal.

[0142] The filter determination unit 6 estimates the type of noise from the input external sound signal. For example, the filter determination unit 6 estimates the type of noise by inputting the external sound signal into a pre-trained noise estimation model.

[0143] The filter determination unit 6 may also estimate the type of noise without using an external sound signal. For example, the filter determination unit 6 may be input with information about the current location of the noise-emitting device. From this information, the filter determination unit 6 estimates the type of noise at the current location of the noise-emitting device. For example, if the input current location of the noise-emitting device is inside a train, the filter determination unit 6 can estimate that the noise is from inside a train.

[0144] The filter determination unit 6 determines a set of filters according to the estimated type of noise. For example, if the type of noise is low-frequency dominant noise such as noise inside a train, airplane, or car, a set of bandpass filters with a low cutoff frequency (H wf 1 (z), H tw 1 It is decided to use (z). Also, in the case of noise types where high frequencies are dominant, such as noise in urban areas or noise inside shops, a set of bandpass filters with a high cutoff frequency (H wf 2 (z), H tw 2 We decide to use (z)).

[0145] The first filter processing unit 11 and the third filter processing unit 21 are the first bandpass filter H wf (z) is the bandpass filter H determined by the filter determination unit 6. wf 1 Perform processing using (z).

[0146] The second filter processing unit 12 and the fourth filter processing unit 22 are second bandpass filters H tw (z) is the bandpass filter H determined by the filter determination unit 6. wf 2 Perform processing using (z).

[0147] In the case of noise dominated by low frequencies, such as noise inside trains, airplanes, and cars, a set of bandpass filters with a low cutoff frequency (H wf 1 (z), H tw 1 It is preferable to use (z). Also, in the case of noise where high frequencies are dominant, such as noise in urban areas or noise inside stores, a set of bandpass filters with a high cutoff frequency (H wf 2 (z), H tw 2 It is preferable to use (z)).

[0148] The above process allows for the use of an appropriate bandpass filter based on the external audio signal.

[0149] (Example 3) A set of bandpass filters that primarily suppress low frequencies (H wf (z), H tw 1 (z)) and a pair of bandpass filters that perform normal suppression (H wf (z), H tw 2 (z)) is predetermined.

[0150] As shown in Figure 10(c), the bandpass filter H wf (z) is the cutoff frequency f c1 This is a 700Hz low-pass filter, and a band-pass filter H tw 1 (z) is the cutoff frequency f c1 ,f c2 These are 700Hz and 1500Hz bandpass filters, respectively. In other words, bandpass filter H tw 1 (z) is a bandpass filter that primarily passes frequencies in the 700Hz to 1500Hz range.

[0151] As shown in Figure 10(d), the bandpass filter H wf (z) is the cutoff frequency f c2 This is a 700Hz low-pass filter, and a band-pass filter H tw 2 (z) is the cutoff frequency f c1 ,f c2 These are 700Hz and 4000Hz bandpass filters, respectively. In other words, bandpass filter H tw 2 (z) is a bandpass filter that primarily passes frequencies in the 700Hz to 4000Hz range.

[0152] The filter determination unit 6 receives an external audio signal. Examples of external audio signals are signals x(n), r(n), and e(n). Signals obtained from other microphones may also be used as the external audio signal.

[0153] The filter determination unit 6 calculates an index value representing the resemblance of ambient sound to speech from the input external sound signal. The following explanation will use an example where a larger index value indicates a higher degree of resemblance to ambient sound.

[0154] The filter determination unit 6 determines whether the calculated value is greater than a predetermined threshold.

[0155] If the calculated value is greater than a predetermined threshold, the filter determination unit 6 determines a set of bandpass filters that primarily suppress low frequencies (H wf (z), H tw 1 It is decided to use (z). Otherwise, the filter determination unit 6 determines the set of bandpass filters (H) that perform normal suppression. wf (z), H tw 2 We decide to use (z)).

[0156] Thus, when there is a high probability that ambient noise includes speech, a set of bandpass filters that primarily suppress low frequencies (H wf (z), H tw 1 By deciding to use (z), it is possible to avoid silencing the voice.

[0157] Therefore, the above process allows for the use of an appropriate bandpass filter based on the external audio signal.

[0158] [Eighth Embodiment] The sound emission device and method of the eighth embodiment will be described below. The explanation will focus on the parts that differ from the first to seventh embodiments. Parts that are the same as those of the first to seventh embodiments will be omitted.

[0159] In the first to seventh embodiments, the filter determination unit 6 may determine a bandpass filter (for example, at least one of the bandpass filters used in the first to sixth filter processing units 11 to 51) based on the sound signal played back from the earphones.

[0160] For example, the filter determination unit 6 may change parameters such as the coefficients and cutoff frequency of the bandpass filter based on the sound signal played back by the earphones. More specifically, the filter determination unit 6 may change these parameters based on the characteristics of the sound signal played back by the earphones. The characteristics of the sound signal played back by the earphones are, for example, the magnitude, power, and spectral centroid of the sound signal played back by the earphones.

[0161] In the following description, as an example of the eighth embodiment of the sound emission device, we will explain an example in which the filter determination unit 6 of the sound emission device of the first embodiment determines a bandpass filter based on the sound output signal from the earphones.

[0162] As illustrated in Figure 11, the eighth embodiment of the sound emission device further comprises a filter determination unit 6, addition units 71, 72, a seventh filter processing unit 81, an eighth filter processing unit 82, and subtraction units 91, 92.

[0163] The audio signal played back by the earphones wf (n) is input to the filter determination unit 6, the summing unit 71, and the seventh filter processing unit 81. The playback sound signal s from the earphones tw (n) is input to the filter determination unit 6, the addition unit 72, and the eighth filter processing unit 82.

[0164] The addition unit 71 adds the first acoustic signal y wf (n) and the audio signal s from the earphones. wf (n) is added. The added signal is output to the first driver unit 41. The adder 72 also adds the second acoustic signal y tw (n) and the audio signal s from the earphones. tw (n) and (n) are added together. The added signal is output to the second driver unit 42. The first driver unit 41 and the second driver unit 42 emit a sound corresponding to the added signal that was input.

[0165] The seventh filter processing unit 81 processes the playback sound signal s from the earphones. wf For (n), the characteristic S from the DA conversion of the first driver unit 41 to the AD conversion of the error microphone EM wfFiltering is performed using a filter that simulates (z). The filtered signal is output to the subtraction unit 91. The eighth filter processing unit 82 processes the earphone playback sound signal s tw For (n), the characteristic S from the DA conversion of the second driver unit 42 to the AD conversion of the error microphone EM tw A filter is applied using a filter that simulates (z). The filtered signal is output to the subtraction unit 92.

[0166] The subtraction unit 91 subtracts the signal filtered by the seventh filter processing unit 81 from the error signal e(n) converted by the error microphone EM. The subtraction unit 92 subtracts the signal filtered by the eighth filter processing unit 82 from that signal. The subtracted signal is output to the third filter processing unit 21 and the fourth filter processing unit 22.

[0167] Thus, the sound signal played back by the earphones wf (n), s tw By canceling (n) through feedback, adaptive filtering can be performed stably.

[0168] By the way, a set of bandpass filters with a high cutoff frequency (H wf 1 (z), H tw 1 (z)) and a set of bandpass filters with a low cutoff frequency (H wf 2 (z), H tw 2 (z)) is predetermined.

[0169] As shown in Figure 12(a), the bandpass filter H wf 1 (z) is the cutoff frequency f c1 This is a 700Hz low-pass filter, and a band-pass filter H tw 1 (z) is the cutoff frequency f c1 This is a 700Hz high-pass filter. As shown in Figure 12(b), the band-pass filter H wf 2 (z) is the cutoff frequency f c2This is a 300Hz low-pass filter, and a band-pass filter H tw 2 (z) is the cutoff frequency f c2 This is a 300Hz high-pass filter.

[0170] In this case, the filter determination unit 6 determines whether or not there is an earphone playback sound signal. For example, the filter determination unit 6 determines that there is an earphone playback sound signal if the power of the earphone playback sound signal is above a predetermined threshold.

[0171] If it is determined that there is an earphone playback sound signal, the filter determination unit 6 determines a set of bandpass filters with a low cutoff frequency (H wf 2 (z), H tw 2 It is decided to use (z). Otherwise, the filter determination unit 6 determines a set of bandpass filters with a high cutoff frequency (H wf 1 (z), H tw 1 We decide to use (z)).

[0172] When there is an audio signal playing through earphones, there is little need for noise reduction in the frequency range where the earphone audio signal masks the noise. Therefore, when there is an audio signal playing through earphones, it is effective to focus noise reduction on the frequency range where the earphone audio signal does not mask the noise, in other words, on low frequencies with relatively high energy.

[0173] Therefore, as described above, when it is determined that there is an earphone playback sound signal, a set of bandpass filters with a low cutoff frequency that suppresses low frequencies (H wf 2 (z), H tw 2 By deciding to use (z), effective sound suppression can be achieved.

[0174] [Variations] The specific configuration of the embodiments of the disclosed technology is not limited to the configuration described above. The specific configuration of the embodiments of the disclosed technology can be modified as appropriate, without departing from the spirit of the embodiments of the disclosed technology.

[0175] The noise emission device may employ active noise control methods other than the feedforward method. Examples of other active noise control methods include, for example, the feedback method and the hybrid method.

[0176] As mentioned earlier, the second driver unit 42 may be a smaller driver unit than the first driver unit. In this case, the first filter processing unit 11 uses a band-limiting filter H wf Without performing filtering using (z), - S wf Only filtering using (z) may be performed. Also, the third filter processing unit 21 does not have to be provided in the sound emission device. In this case, the noise cancellation unit 3 uses the first acoustic signal ~r based on the noise-based acoustic signal wf It can be said that (n) is controlled to emit sound from the first driver unit 41.

[0177] Bandwidth limiting filter H tw When performing the (z) process, the bandwidth that increases delay can be brought to a bandwidth that can be eliminated by, for example, the first driver unit 41 which is a woofer, thereby minimizing the impact.

[0178] The low-frequency components of ambient noise energy tend to be large. Therefore, the second driver unit, such as the tweeter, may also suppress low frequencies, potentially due to the low-frequency components that the first driver unit 41 failed to eliminate. In this case, if the noise pressure level is high, distortion may occur if there is no high-pass filter directly before the second driver unit 42, potentially causing the ANC (Active Noise Cancellation) to malfunction.

[0179] Therefore, the first filter processing unit 11 performs a bandwidth limiting filter H wf Without performing filtering using (z), - S wf Even when only filtering using (z) is performed, the second filter processing unit 12 performs the band limiting filter H swFiltering using (z) may be performed. The various processes described in the embodiments of the disclosed technology may be performed not only in chronological order according to the order described, but also in parallel or individually as required by the processing capacity of the device performing the processes. For example, steps S11, S12, S21, and S22 may be processed in parallel.

[0180] For example, data exchange between components of a sound emission device may occur directly, or it may occur via a storage unit not shown in the diagram.

[0181] Furthermore, the present invention may also include a device (terminal) for using the apparatus, system, or method of the present invention via a network (telecommunication line). The "device (terminal) for use" may be equipped with functions necessary to obtain the effects of implementing the apparatus, system, or method of the present invention (for example, control functions, decoding functions, restoration functions, input / output functions, etc.).

[0182] It goes without saying that the invention may be modified as appropriate without departing from its spirit.

[0183] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually described as being incorporated by reference.

[0184] [Experimental Example] First, let's explain the conditions of the experiment.

[0185] The noise source used was a 20-second recording of noise inside an airplane. For the secondary sound source, the primary path (from the noise source to the reference microphone) and the secondary path (from the secondary sound source to the error microphone), using the woofer and tweeter of the nwm one (registered trademark), were measured using exponential chirp signals. A hybrid time-domain FxLMS algorithm was used as the ANC (Active Noise Cancellation) algorithm. Two reference microphones were used, positioned on the front and back of the headphones. One error microphone was used, positioned at the entrance of the auricle when the headphones were worn.

[0186] For evaluation, the impulse response from a primary sound source (noise speaker) and a secondary sound source (open-ear headphones) was measured for each microphone, and ANC (Active Noise Cancellation) was performed using simulation for evaluation. A 4mm diameter ECM (electret condenser microphone) was used as the microphone.

[0187] Furthermore, the primary sound source is assumed to be positioned in front of the dummy head wearing open-ear headphones, which are the secondary sound source.

[0188] First, we simulate and calculate the observed signal (noise + secondary sound source) without ANC and with ANC applied.

[0189] Then, the Short-Time Fourier Transform (STFT) is used to convert each observed signal into the time-frequency domain.

[0190] Identify the section where the ANC noise control filter has sufficiently converged and effectively suppresses noise. To evaluate the noise suppression effect, calculate the average value L(f) of the difference in the logarithmic power spectrum in that section. Specifically, use data from approximately 5 seconds (250 frames in total, 1024 samples per frame, sampling frequency = 48 kHz, 1 / 4 overlap) from the end of the observed signal.

[0191] A larger value of this difference L(f) indicates higher noise suppression performance at frequency f.

[0192] Furthermore, in order to approximate the performance of an actual DSP, an artificial delay equivalent to AD-DA conversion is introduced into the secondary path. In Figure 13, the dashed line TW is H tw (z) is used to show the amount of noise suppression when ANC processing is applied to the tweeter alone. In Figure 13, the dotted line WF shows the amount of noise suppression when ANC processing is applied to the woofer alone. In Figure 13, the solid line WF+TW shows the amount of noise suppression when ANC processing is applied to both the tweeter and the woofer.

[0193] Figure 13 shows that the woofer alone performs well below approximately 2500Hz. However, above 2500Hz, the tweeter alone performs better. Furthermore, it can be seen that using both the woofer and tweeter as secondary sound sources improves noise suppression performance across the entire frequency range up to 4000Hz.

[0194] In Figure 13, the first filter processing unit 11 to the fourth filter processing unit 22 do not perform processing using the first bandpass filter and the second bandpass filter. Even in this case, it can be seen that more efficient noise suppression performance can be obtained by using two drive units of different sizes.

[0195] In Figure 14, the solid line WF+TW shows the amount of noise suppression when ANC processing is applied to the tweeter and woofer. In Figure 14, the dashed lines WF:HP80,TW:HP80 are H wf (z) is a second-order Butterworth filter, which is an 80Hz high-pass filter, H tw (z) represents the amount of noise suppression when ANC processing is applied to the tweeter and woofer using a second-order Butterworth filter, which is an 80Hz high-pass filter. In Figure 14, the dotted lines WF:HP80, TW:HP200 are H wf (z) is a second-order Butterworth filter, which is an 80Hz high-pass filter, H tw The noise suppression amount when ANC processing is applied to the tweeter and woofer using a second-order Butterworth filter, which is a 200Hz high-pass filter, is shown as (z). In Figure 14, the solid lines WF:HP80,LP1000, TW:HP1000,LP2500 are H wf As (z), a filter consisting of a second-order Butterworth filter, which is an 80Hz high-pass filter, and a second-order Butterworth filter, which is a 1000Hz low-pass filter, connected in series, H tw (z) represents the amount of noise suppression when ANC processing is applied to the tweeter and woofer using a filter consisting of a second-order Butterworth filter (a 1000Hz high-pass filter) and a second-order Butterworth filter (a 2500Hz low-pass filter) connected in series.

[0196] Figure 14 shows that noise suppression performance can be improved by applying frequency band limiting filters suitable for each speaker unit and placement to the woofer and tweeter.

[0197] [Programs, Recording Media] The functions realized by the components described herein may be implemented in a circuitry or processing circuitry, including a general-purpose processor, an application-specific processor, an integrated circuit, an ASIC (Application Specific Integrated Circuit), a CPU (a Central Processing Unit), conventional circuits, and / or a combination thereof, programmed to realize the functions described herein. A processor includes transistors and other circuits and is considered a circuitry or processing circuitry. A processor may be a programmed processor that executes a program stored in memory.

[0198] In this specification, circuitry, unit, and means are hardware programmed to perform or execute the functions described herein. Such hardware may be any hardware disclosed herein, or any hardware known to be programmed to perform or execute the functions described herein.

[0199] If the hardware is a processor that is considered to be a type of circuitry, then the circuitry, means, or unit is a combination of hardware and software used to constitute the hardware and / or processor.

[0200] The various processes described above can be carried out by loading a program that executes each step of the above method into the recording unit 2020 of the computer 2000 shown in Figure 16, and then causing the control unit 2010, input unit 2030, output unit 2040, display unit 2050, etc. to operate.

[0201] The program describing this process can be recorded on a computer-readable recording medium. Any computer-readable recording medium can be used, such as a magnetic recording device, optical disc, magneto-optical recording medium, or semiconductor memory.

[0202] Furthermore, this program may be distributed, for example, by selling, transferring, or lending portable recording media such as DVDs or CD-ROMs on which the program is recorded. Alternatively, the program may be stored in the storage device of a server computer and distributed by transferring the program from the server computer to other computers via a network.

[0203] A computer executing such a program may, for example, first store the program recorded on a portable storage medium or a program transferred from a server computer in its own storage device. Then, when processing is to be executed, the computer reads the program stored on its own storage medium and executes the processing according to the read program. Alternatively, the computer may directly read the program from the portable storage medium and execute the processing according to that program, or it may sequentially execute the processing according to the received program each time a program is transferred to it from a server computer. Furthermore, the processing may be executed using a so-called ASP (Application Service Provider) type service, where the processing function is realized only by issuing execution instructions and obtaining results, without transferring the program from the server computer to this computer.In addition, the processing may be executed using a so-called SaaS (Software as a Service) type service, where a part of the server computer is made available to the user along with the program. Furthermore, the term "program" in this form includes information used for processing by an electronic computer that is equivalent to a program (data, etc., that is not a direct instruction to the computer but has the property of defining the processing of the computer).

[0204] Furthermore, in this configuration, the device is configured by executing a predetermined program on a computer, but at least a part of these processes may be implemented in hardware.

[0205] [Additional Embodiments] As previously mentioned, the sound emission device may be a glasses-type sound emission device such as audio glasses or smart glasses. In this case, as illustrated in Figures 17 and 18, the first driver unit 41 and the second driver unit 42 may be arranged on the temple 101 of the glasses-type sound emission device. If a protrusion 1011 is provided on the temple 101, as illustrated in Figure 18, at least one of the first driver unit 41 and the second driver unit 42 may be arranged on the protrusion 1011 of the temple 101. In the example of Figure 18, the second driver unit 42 is arranged on the protrusion 1011 of the temple 101.

[0206] In the following explanation, the first driver unit 41 is intended to cancel low-frequency components, and the second driver unit 42 is intended to cancel components other than low-frequency components. Low-frequency components are frequency components below a predetermined frequency.

[0207] In other words, the first signal corresponding to the first acoustic signal emitted from the first driver unit 41 may be a signal that cancels out the low-frequency components of the external noise, and the second signal corresponding to the second acoustic signal emitted from the second driver unit 42 may be a signal that cancels out the external noise other than the low-frequency components.

[0208] For this reason, the diaphragm of the first driver unit 41 may be larger than the diaphragm of the second driver unit 42. In other words, the lowest resonant frequency of the first driver unit 41 may be less than or equal to the lowest resonant frequency of the second driver unit 42.

[0209] In the examples shown in Figures 17 and 18, the distance between the first driver unit 41 and the hinge 102 is shorter than the distance between the second driver unit 42 and the hinge 102. In other words, the distance between the second driver unit 42 and the entrance to the ear canal is shorter than the distance between the first driver unit 41 and the entrance to the ear canal.

[0210] This makes the secondary path delay of the second driver unit 42 smaller than that of the first driver unit 41, thereby enabling efficient cancellation of high-frequency components.

[0211] Furthermore, since high-frequency components have strong directivity, a sealed speaker may be used as the second driver unit 42. A sealed speaker restricts the movement of the speaker unit's diaphragm, making it easier to concentrate the direction of sound wave radiation. Therefore, by using a sealed speaker, performance that takes advantage of the directivity of high-frequency components can be obtained.

[0212] On the other hand, a dipole (inverse phase radiating) speaker may be used as the first driver unit 41. This makes it possible to take advantage of the low-frequency characteristics (i.e., a wide radiation range).

[0213] Of course, both the first driver unit 41 and the second driver unit 42 may be dipole type (reverse phase radiating type) speakers. The type of speakers used for the first driver unit 41 and the second driver unit 42 will be appropriately selected according to the required performance.

[0214] As mentioned earlier, the first driver unit 41 and the second driver unit 42 may be two driver units with different characteristics included in a 2-way coaxial speaker. For example, as shown in Figure 19, the diaphragm 411 of the first driver unit 41 may be ring-shaped, and the diaphragm 421 of the second driver unit 42 may be circular, and the diaphragm 411 of the first driver unit and the diaphragm 421 of the second driver unit 42 may be arranged on the same plane. Figure 19 shows the positional relationship between the diaphragm 411 of the first driver unit 41 and the diaphragm 421 of the second driver unit 42, and therefore other components of the first driver unit 41 and the second driver unit 42 are omitted.

[0215] As illustrated in Figure 20, the first driver unit 41 and the second driver unit 42, which are arranged coaxially, may be housed in a sealed enclosure.

[0216] In this case, as a pseudo-secondary path characteristic, the characteristics from the center of the second driver unit 42 to a microphone M located 1 cm or 10 cm away were measured in a free sound field. Specifically, the group delay was measured for both the first driver unit 41 and the second driver unit 42. The group delay is the sum of the mechanical group delay of each speaker unit, the acoustic group delay due to the speaker position, and the group delay of the electrical system. The outer diameter of the first driver unit 41 used in the measurement was 74 mm, the thickness was 16.2 mm, and the inner diameter was 35 mm. The outer diameter of the second driver unit 42 used in the measurement was 35 mm, and the thickness was 9.5 mm.

[0217] Figure 21 shows the group delay measured with microphone M located 1 cm away from the center of the second driver unit 42. Figure 22 shows the group delay measured with microphone M located 10 cm away from the center of the second driver unit 42. In Figures 21 and 22, the horizontal axis represents frequency and the vertical axis represents group delay.

[0218] Figures 21 and 22 show that as the listening point (in other words, the position of microphone M) approaches the first driver unit 41 and the second driver unit 42, the group delay of the second driver unit 42 becomes smaller than the group delay of the first driver unit 41. This can be attributed to the fact that, as explained below, the difference in acoustic group delays increases as the listening point approaches the first driver unit 41 and the second driver unit 42.

[0219] As illustrated in Figure 20, let r be the radius of the first driver unit 41, d1 be the distance between the first driver unit 41 and the microphone M, and d2 be the distance between the second driver unit 42 and the microphone M. In this case, by the Pythagorean theorem, d1 2 =d2 2 +r 2 It can be considered that way.

[0220] r≪d 2 In this case, that is, when the listening point is sufficiently far from the first driver unit 41 and the second driver unit 42, d1 2 ≒d2 2 In this case, the difference in acoustic group delay between the first driver unit 41 and the second driver unit 42 can be almost ignored.

[0221] On the other hand, the listening point is near the first driver unit 41 and the second driver unit 42, and r≈d2 2 At that time, d1 2 ≒2d2 2 In this case, the difference in acoustic group delay between the first driver unit 41 and the second driver unit 42 is large.

[0222] For example, if r = 50 mm and d2 = 100 mm, then d1 = (d2 2 +r 2 ) 1 / 2 This is approximately 112.99 mm. At this time, the difference in distance between d2 and d1 is approximately 12 mm. Assuming the speed of sound is 340 m / s, the difference in acoustic group delay is 0.012 / 340 ≈ 35 microseconds.

[0223] On the other hand, for example, if r = 50 mm and d2 = 10 mm, then d1 = (d2 2 +r2 ) 1 / 2 This is approximately 50.99 mm. At this time, the difference in distance between d2 and d1 is approximately 41 mm. Assuming the speed of sound is 340 m / s, the difference in acoustic group delay is 0.041 / 340 ≈ 120 microseconds. This means that the sound waves emitted from the first driver unit 41 are delayed by approximately 120 microseconds compared to the sound waves emitted from the second driver unit 42.

[0224] From this, it can be seen that when the listening point is near the first driver unit 41 and the second driver unit 42, the second driver unit 42 is more suitable than the first driver unit 41 for high-frequency noise control that is strict about delay.

[0225] Thus, when the first driver unit 41 and the second driver unit 42, which are arranged coaxially, are housed in a sealed enclosure, it can be considered that the acoustic group delay is determined by the distance between the listening point and the first driver unit 41 and the second driver unit 42.

[0226] On the other hand, if the first driver unit 41 and the second driver unit 42, which are arranged coaxially, are housed in an open-type enclosure rather than a sealed type, the radiation characteristics of the first driver unit 41 and the second driver unit 42 become dipole type (radiating sound waves in opposite phase), and the effect of the phase characteristics due to the proximity effect between the speaker and microphone becomes more pronounced. Due to this difference in phase characteristics, the difference in acoustic group delay becomes larger than in the case of a sealed enclosure.

[0227] The first driver unit 41 and the second driver unit 42 are two driver units with different characteristics included in a 2-way coaxial speaker, and the diaphragm 411 of the first driver unit and the diaphragm 421 of the second driver unit 42 do not have to be arranged on the same plane. For example, the second driver unit 42 and its diaphragm 421 may be positioned closer to the ear and the entrance to the ear canal than the first driver unit 41 and its diaphragm 411.

[0228] The first driver unit 41, which is a woofer, may be ring-shaped. Alternatively, the second driver unit 42 may be a tweeter, and when the first driver unit 41 is projected onto a predetermined plane, the second driver unit 42 may be included within the ring-shaped first driver unit 41. The predetermined plane is a plane parallel to the ring-shaped first driver unit 41. For example, such a first driver unit 41 and second driver unit 42 may be arranged in the nwm one® as illustrated in Figures 23 and 24. Figure 23 shows the external appearance of the nwm one®. Figure 24 shows the interior of the sound-emitting section 40 of the nwm one®. As shown in Figure 24, inside the sound-emitting section 40, the first driver unit 41, which is a woofer, and the second driver unit 42, which is a tweeter, are arranged coaxially, and the second driver unit 42 and its diaphragm 421 are positioned closer to the ear and ear canal entrance than the first driver unit 41 and its diaphragm 411.

[0229] In this case, as illustrated in Figure 24, the error microphone EM may be positioned closest to the ear / external auditory canal entrance within the sound-emitting section 40. As a result, the distance between the error microphone EM and the first driver unit 41 may be longer than the distance between the error microphone EM and the second driver unit 42. Also, as shown in Figure 23, the reference microphone RM may be positioned at least one of the following locations: the center P1 on the back of the sound-emitting section 40, the front position P2 of the annular housing holding the sound-emitting section 40, and the rear position P3 of the annular housing holding the sound-emitting section 40. Front and rear refer to the front and rear of the user's body when the user is wearing the nwm one (registered trademark).

[0230] As shown in Figure 24, with the error microphone EM positioned closest to the ear / external auditory canal entrance within the sound emission unit 40, the (1) relative logarithmic power spectrum and (2) group delay were measured by simulation as secondary path characteristics from the first driver unit 41 and the second driver unit 42 to the error microphone EM. The measurement results are shown in Figures 25 and 26. Note that, as this is a simulation, electrical characteristics and delays are not included. In Figures 25 and 26, the solid line shows the value for the second driver unit 42, and the dashed line shows the value for the first driver unit 41. The horizontal axis in Figures 25 and 26 is frequency. The vertical axis in Figure 25 is the relative logarithmic power spectrum. The vertical axis in Figure 26 is the mechanical and acoustic group delay.

[0231] Figure 25 shows that because the diaphragm of the second driver unit 42, which is a tweeter, is small, the lowest resonant frequency (F0) of the second driver unit 42 is higher than that of the first driver unit 41, which is a woofer. Furthermore, when viewed over a wide bandwidth, it can be seen that the efficiency of the second driver unit 42, which is a tweeter, tends to be low.

[0232] When attempting to suppress high-sound-pressure noise in a frequency range where the speaker unit's efficiency is low, nonlinear distortion occurs, degrading the noise suppression performance. Therefore, it is expected that it will be difficult to suppress low-frequency or high-sound-pressure noise with a tweeter alone.

[0233] Figure 26 shows that the group delay of the second driver unit 42, which is a tweeter, is large below 1 kHz. On the other hand, the group delay of the first driver unit 41, which is a woofer, is large above 2 kHz. Therefore, from the perspective of group delay, it can be seen that the woofer is suitable for noise suppression below 1 kHz, and the tweeter is suitable above 2 kHz.

[0234] When a glasses-type sound emission device such as audio glasses or smart glasses is used, the first driver unit 41 and the second driver unit 42 may be provided in a single housing, as shown in Figures 27 and 28.

[0235] Figure 27 shows a case where the first driver unit 41 and the second driver unit 42 are sealed. Small ventilation holes may be provided on the backs of the first driver unit 41 and the second driver unit 42. In the example of Figure 27, the first driver unit 41 and the second driver unit 42 are fixed to the housing by a partition plate 43 inside the housing. The partition plate 43 also provides an air chamber 412 on the positive phase side of the first driver unit 41 and an air chamber 422 on the positive phase side of the second driver unit 42.

[0236] The air chamber 412 is provided with a first sound hole 413. The air chamber 422 is provided with a second sound hole 423. The first sound hole 413 is a sound hole that releases sound emitted from the first driver unit 41 to the outside. The second sound hole 423 is a sound hole that releases sound emitted from the second driver unit 42 to the outside. In this example, the first sound hole 413 is located on the upper side of the temple 101 relative to the ground. The second sound hole 423 is located on the lower side of the temple 101 relative to the ground.

[0237] In the example shown in Figure 27, the distance between the first sound hole 413 and the entrance to the ear canal is longer than the distance between the second sound hole 423 and the entrance to the ear canal. Therefore, the acoustic group delay of the first driver unit 41 is greater than the acoustic group delay of the second driver unit 42.

[0238] By separating the positive-phase air chamber 412 of the first driver unit 41 from the positive-phase air chamber 422 of the second driver unit 42, the volumes of these air chambers 412 and 422 can be reduced. This allows for a higher resonant frequency, preventing deterioration of the sound quality and ANC performance of the reproduced sound.

[0239] Figure 28 shows the case where the first driver unit 41 and the second driver unit 42 are open-type (in other words, dipole type). In the example of Figure 28, the first driver unit 41 and the second driver unit 42 are fixed to the housing by a partition plate 43 inside the housing. The partition plate 43 also provides an air chamber 441 on the positive phase side of the first driver unit 41 and the second driver unit 42, and an air chamber 451 on the negative phase side of the first driver unit 41 and the second driver unit 42. The positive phase air chamber 441 is provided with a positive phase radiating hole 442, and the negative phase air chamber 451 is provided with a negative phase radiating hole 452. In the example of Figure 28, the distance between the first driver unit 41 and the entrance to the ear canal is longer than the distance between the second driver unit 42 and the entrance to the ear canal. Therefore, in the example of Figure 28, the acoustic group delay of the first driver unit 41 is greater than the acoustic group delay of the second driver unit 42.

Claims

1. A sound-emitting device worn by a user, comprising: a first driver unit and a second driver unit, wherein the sound-emitting device is configured such that noise arriving from the outside can reach the user's ears, the sound emitted from the first driver unit is controlled to be a first acoustic signal based on a first signal obtained by applying a first bandpass filter to an acoustic signal based on noise, the sound emitted from the second driver unit is controlled to be a second acoustic signal based on a second signal obtained by applying a second bandpass filter to an acoustic signal based on noise, and the diaphragm of the first driver unit is larger than the diaphragm of the second driver unit.

2. A sound emission device according to claim 1, wherein the sound emission device generates an adaptively updated filter, and the filter is generated from the first signal and a signal obtained by applying the first bandpass filter to the error signal.

3. A sound emission device according to claim 1, further comprising a filter determination unit that determines the first bandpass filter and the second bandpass filter based on an external sound signal.

4. The sound emission device according to any one of claims 1 to 3, further comprising an error microphone, wherein the distance between the error microphone and the first driver unit is longer than the distance between the error microphone and the second driver unit.

5. The sound emission device according to claim 4, wherein the first signal is a signal that cancels the low-frequency component of the noise, and the second signal is a signal that cancels the components of the noise other than the low-frequency component.

6. The sound emitting device is a spectacle-type sound emitting device, the first driver unit and the second driver unit are arranged on the temple, and the distance between the first driver unit and the hinge is shorter than the distance between the second driver unit and the hinge, according to claim 1.

7. The sound-emitting device according to claim 6, further comprising: a first sound hole for emitting sound emitted from the first driver unit to the outside; and a second sound hole for emitting sound emitted from the second driver unit to the outside, wherein the first sound hole is positioned on the upper side of the temple with respect to the ground, and the second sound hole is positioned on the lower side of the temple with respect to the ground.

8. The sound emission device according to claim 1, wherein the diaphragm of the first driver unit is ring-shaped, the diaphragm of the second driver unit is circular, and the diaphragm of the first driver unit and the diaphragm of the second driver unit are arranged on the same plane.

Images