Audio signal processing device and remote control system

Abstract

The present invention addresses the problem of providing an audio signal processing device and a remote control system for removing noise components in an audio signal with a simple configuration. [Solution] An IFFT (232) outputs, to an adaptive filter (233), a reference sound generated by inverse fast Fourier transform of Fourier coefficients in noise element sound data (231). A subtractor (234) subtracts the noise cancelled sound output by the adaptive filter from the audio signal (S) and outputs the noise cancelled sound as an output signal (S ') to a speaker (24), and the adaptive filter performs filter coefficient adaptive operation with the output signal (S') as an error. In the noise element sound data, Fourier coefficients representing information on the frequency domain of a synthesized tone of pure tones of a plurality of frequencies extracted at predetermined frequency intervals from within a frequency band to be removed that does not include the main frequency band of a sound component to be listened of an audio signal S but includes a frequency band of an expected noise component are stored.

Description

Technical Field

[0001] The present invention relates to a technique for reducing the level of unknown noise components included in an audio signal. Background Art

[0002] As a technology for reducing the level of unknown noise components included in an audio signal, the following technology is known: using blind source separation to remove the unknown noise components included in the audio signal. In blind source separation, the independence of each sound source signal is utilized to separate each sound source signal from a mixed signal formed by mixing multiple sound source signals (for example, patent documents 1 and 2).

[0003] In this technology, a plurality of microphones are used to collect sound in a noisy space where a sound to be listened to is emitted, and the correlation of audio signals outputted from the respective microphones is evaluated to extract only the sound component to be listened to.

[0004] Prior art literature:

[0005] Patent Literature:

[0006] Patent Document 1: Japanese Patent No. 6966750

[0007] Patent Document 2: Japanese Patent No. 4616529 Summary of the invention

[0008] Problems to be solved by the invention:

[0009] According to the above-mentioned technology of removing unknown noise components by blind source separation, multiple microphones need to be configured, so it is not applicable to the case where it is impossible to ensure the configuration space of multiple microphones in the space where the sound to be listened to is emitted. In addition, even if the configuration space of multiple microphones can be ensured, it will cause complicated operations or increase costs to configure multiple microphones.

[0010] Therefore, an object of the present invention is to reduce the level of an unknown noise component included in an audio signal with a relatively simple configuration.

[0011] Means used to solve the problem:

[0012] In order to achieve the above-mentioned problem, the present invention is an audio signal processing device that suppresses the noise component of an input signal including a listening object sound component and a noise component and outputs it as an output signal, comprising: a synthetic sound output mechanism that outputs a synthetic sound signal; an adaptive filter that takes the synthetic sound signal output by the synthetic sound output mechanism as input; and a subtraction mechanism that subtracts the output signal of the adaptive filter from the input signal and outputs it as the output signal. Here, the adaptive filter performs adaptive operation using the output signal output by the subtraction mechanism as an error. In addition, the synthetic sound output mechanism outputs a signal of a synthetic sound of pure tones of multiple frequencies extracted from a removal object frequency band as the synthetic sound signal, and the removal object frequency band is a frequency band of the noise component that is the suppression object in the audio signal processing device.

[0013] Here, in the audio signal processing device, the synthetic sound output means may output, as the synthetic sound signal, a signal of a synthetic sound of pure tones of a plurality of frequencies extracted at regular frequency intervals from within the removal target frequency band.

[0014] Here, the audio signal processing device may be a synthetic sound output device composed of an information storage device and a conversion device, wherein the information storage device stores frequency domain information of the synthetic sound signal to be output, and the conversion device converts the information stored in the information storage device into the synthetic sound signal.

[0015] In this case, in the audio signal processing device, the frequency domain information of the synthetic audio signal may be used as Fourier coefficients, and the conversion means may perform conversion into the synthetic audio signal by inverse fast Fourier transform.

[0016] Alternatively, in this case, in the audio signal processing device, the frequency domain information of the synthetic audio signal may be used as wavelet transform coefficients, and the conversion means may perform conversion into the synthetic audio signal by inverse wavelet transform.

[0017] In addition, in order to achieve the above-mentioned problem, the present invention is an audio signal processing device that suppresses the noise component of an input signal including a sound component of a listening object and a noise component and outputs it as an output signal, comprising: a first conversion mechanism that converts the input signal into first information that is information in the frequency domain of the input signal; an information storage mechanism that stores information in the frequency domain of the synthetic sound signal as second information, a signal of a synthetic sound of a plurality of pure sounds extracted from a frequency band to be removed as a synthetic sound signal, the frequency band to be removed being a frequency band of a noise component to be suppressed in the audio signal processing device; an adaptive filter that takes the second information stored in the information storage mechanism as input and outputs third information that is information in the frequency domain of the noise-canceled sound; a subtraction mechanism that outputs information in the frequency domain obtained by subtracting the third information from the first information as fourth information; and a second conversion mechanism that converts the fourth information into a sound signal and outputs it as the output signal. Here, the adaptive filter performs adaptive operation by treating the fourth information as an error.

[0018] Here, in such an audio signal processing device, the synthesized audio signal may be a signal of a synthesized sound of pure tones of a plurality of frequencies extracted at regular frequency intervals from the removal target frequency band.

[0019] In addition, in such an audio signal processing device, the frequency domain information can also be used as Fourier coefficients, and in the first conversion mechanism, a fast Fourier transform is performed to convert the input signal into the first information, and in the second conversion mechanism, an inverse fast Fourier transform is performed to convert the fourth information into the output signal.

[0020] Alternatively, in such an audio signal processing device, the frequency domain information can also be used as wavelet transform coefficients, and in the first conversion mechanism, wavelet transform is performed to convert the input signal into the first information, and in the second conversion mechanism, inverse wavelet transform is performed to convert the fourth information into the output signal.

[0021] In addition, at the same time, the present invention also provides a remote control system having the above audio signal processing device. The remote control system has a vehicle-side system mounted on a vehicle, and a control-side system that remotely controls the vehicle, the vehicle-side system having: a microphone that picks up a sound mixed with the listening object sound and the noise and outputs it as a monitoring sound signal; and a sending mechanism that sends the monitoring sound signal output by the microphone to the control-side system. In addition, the control-side system has a receiving mechanism that receives the monitoring sound signal from the vehicle-side system, a speaker, and the audio signal processing device, the monitoring sound signal received by the receiving mechanism is input to the audio signal processing device as the input signal, and the output signal is output from the audio signal processing device to the speaker.

[0022] According to the above audio signal processing device, with a simple configuration that does not require a plurality of microphones, it is possible to reduce the level of an unknown noise component included in an input signal and output the signal as an output audio signal.

[0023] Effect of the invention:

[0024] As described above, according to the present invention, the level of the unknown noise component included in the audio signal can be reduced with a relatively simple configuration. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a block diagram showing the configuration of a remote control system according to the first embodiment of the present invention.

[0026] Figure 2 This is a block diagram showing the configuration of an audio signal processing device according to the first embodiment of the present invention.

[0027] Figure 3 It is a diagram schematically showing reference audio data according to the first embodiment of the present invention.

[0028] Figure 4 It is a diagram showing an example of noise cancellation sound according to the first embodiment of the present invention.

[0029] Figure 5 This is a block diagram showing another configuration example of the audio signal processing device according to the second embodiment of the present invention.

[0030] Figure 6 This is a block diagram showing another configuration example of the audio signal processing device according to the third embodiment of the present invention.

[0031] Figure 7 This is a block diagram showing another configuration example of the audio signal processing device according to the fourth embodiment of the present invention.

[0032] Figure 8 This is a block diagram showing another configuration example of the audio signal processing device according to the fifth embodiment of the present invention.

[0033] Description of reference numerals:

[0034] 1……Vehicle system, 2……Remote control room system, 3……Network, 11……Actuator, 12……Camera, 13……Sensor, 14……Microphone, 15……Control device, 21……Controller, 22……Monitor, 23……Audio signal processing device, 24……Speaker, 25……Remote control device, 231……Noise element sound data, 232……IFFT, 233……Adaptive filter, 234……Subtractor, 531……Noise element sound data, 532……Frequency domain adaptive filter, 533...FFT, 534...subtractor, 535...IFFT, 631...noise element sound data, 632...inverse wavelet transform unit, 633...adaptive filter, 634...subtractor, 731...noise element sound data, 732...wavelet transform coefficient domain adaptive filter, 733...wavelet transform unit, 734...subtractor, 735...inverse wavelet transform unit, 831...sound source device, 832...adaptive filter, 833...subtractor. DETAILED DESCRIPTION

[0035] The following describes the embodiments of the present invention by taking an example in which the embodiment of the present invention is applied to a remote control system.

[0036] First, the first embodiment will be described.

[0037] Figure 1 The configuration of the remote control system according to the first embodiment is shown.

[0038] The remote control system is a system for remotely controlling a construction vehicle such as a heavy machine, and includes a vehicle system 1 mounted on the construction vehicle and a remote control room system 2 installed in a remote control room.

[0039] The vehicle system 1 includes an actuator 11 for operating the movement of the engineering vehicle, a camera 12 for photographing the surroundings or the interior of the vehicle, a sensor 13 for detecting the status of the engineering vehicle or the surroundings, a microphone 14, and a control device 15. The microphone 14 picks up a sound that is a mixture of an operating sound or a warning sound and noise such as engine noise.

[0040] Then, the control device 15 sends the image captured by the camera 12, the status detected by the sensor 13, or the audio signal picked up by the microphone 14 to the remote control room system 2 via the network 3, and drives the actuator 11 according to the control signal received from the remote control room system 2 to move the engineering vehicle.

[0041] On the other hand, the remote control room system 2 includes a controller 21 , a monitor 22 , an audio signal processing device 23 , a speaker 24 , and a remote control device 25 .

[0042] The remote control device 25 sends a control signal according to the operator's operation received by the controller 21 to the vehicle system 1 via the network 3, thereby controlling the operation of the construction vehicle. In addition, the remote control device 25 displays the image or status received from the vehicle system 1 via the network 3 on the monitor 22, and outputs the audio signal received from the vehicle system 1 via the network 3 to the audio signal processing device 23.

[0043] Then, the audio signal processing device 23 performs signal processing for suppressing noise components on the audio signal output from the remote control device 25 , and outputs the audio signal to the speaker 24 .

[0044] Next, the audio signal processing device 23 will be described.

[0045] Figure 2 The configuration of the audio signal processing device 23 is shown.

[0046] As shown in the figure, the audio signal processing device 23 includes noise element sound data 231, IFFT232 for performing inverse fast Fourier transform, an adaptive filter 233, and a subtractor 234. The subtractor 234 subtracts the output of the adaptive filter 233 from the audio signal S input from the remote control device 25, and outputs it to the speaker 24 as an output signal S'.

[0047] The noise element sound data 231 stores Fourier coefficients, and the IFFT 232 outputs a reference sound generated by performing an inverse fast Fourier transform on the Fourier coefficients to the adaptive filter 233 .

[0048] The adaptive filter 233 applies filtering processing to the input reference sound and outputs it as noise-canceled sound to the subtractor 234. In addition, the adaptive filter 233 uses the output signal S' output by the subtractor 234 as an error, and adapts the filter coefficients (transfer function, tap coefficients) of the filtering processing applied to the reference sound through an adaptive algorithm such as LMS or NLMS so as to minimize the error.

[0049] The Fourier coefficients stored in the noise element sound data 231 are used to express a synthesized sound of pure tones of a plurality of frequencies extracted at predetermined frequency intervals from a removal target frequency band, which is a predetermined frequency band, using information in the frequency domain.

[0050] That is, the Fourier coefficients stored in the noise element sound data 231 are determined so that the reference sound after the inverse fast Fourier transform by the IFFT 232 becomes a composite sound of pure tones of a plurality of frequencies extracted at predetermined frequency intervals from the removal target frequency band.

[0051] Therefore, the noise element sound data 231 and the IFFT 232 function as a sound source device that outputs the synthesized sound of the pure tone.

[0052] The Fourier coefficients stored in the noise element sound data 231 are equivalent to, for example, F(ω) obtained by the following equation 1 using the synthetic sound as f(t).

[0053] In addition, for example, the sound f(t) obtained by performing inverse fast Fourier transform on the Fourier coefficient F(ω) stored in the noise element sound data 231 according to Expression 2 becomes the reference sound.

[0054] [Formula 1]

[0055]

[0056] The values ​​of the Fourier coefficients of the frequencies of the pure tones, that is, the magnitudes (amplitudes or powers) of the pure tones of a plurality of frequencies synthesized into a synthesized sound are all set to the same magnitude. However, the magnitudes may be changed for each frequency.

[0057] The removal target frequency band is set to a frequency band that does not include a main frequency band of a listening target sound component included in the audio signal S but includes an assumed noise component.

[0058] For example, if the noise component envisioned is engine noise, then Figure 3 As shown in the spectrum of the synthesized sound in a, low-order harmonics are also taken into account, the range of 20Hz-800Hz is used as the removal target frequency band, and the frequency interval of pure sound is set to 5Hz.

[0059] In addition, the frequency band to be removed does not necessarily have to be 20 Hz to 800 Hz, and may be a frequency band in another range such as 20 Hz to 1 kHz.

[0060] In addition, you can also Figure 3 As in the case where two removal target bands are set as exemplified in b, a plurality of removal target bands are set.

[0061] According to such an audio signal processing device 23, by adapting the filter coefficient, the noise cancellation sound output by the adaptive filter 233 approximates the magnitude of the frequency component of the audio signal S at each frequency within the removal target frequency band.

[0062] Therefore, if Figure 4 As shown in the case where the frequency band of 20 Hz-1 kHz is set as the removal target frequency band, the noise cancellation sound is similar to the component of the removal target frequency band included in the audio signal S. As a result, in the output signal S' outputted by the subtractor 234 to the speaker 24, the level of the noise component of the audio signal S is reduced.

[0063] On the other hand, since there is no significant impact on frequency bands other than the removal target frequency band of the audio signal S, the target sound component of the audio signal S is not substantially attenuated and is included in the output signal S′, and the operator can well hear the target sound from the output sound of the speaker 24 .

[0064] The first embodiment of the present invention has been described above.

[0065] Next, a second embodiment of the present invention will be described.

[0066] The second embodiment performs signal processing equivalent to the operation of the audio signal processing device 23 of the first embodiment in the frequency domain, and only the configuration of the audio signal processing device 23 is different.

[0067] Figure 5 The configuration of the audio signal processing device 23 according to the second embodiment is shown.

[0068] As shown in the figure, the audio signal processing device 23 according to the second embodiment includes noise element sound data 531 , a frequency domain adaptive filter 532 , an FFT 533 , a subtractor 534 , and an IFFT 535 .

[0069] The noise element sound data 531 stores the same Fourier coefficients as those of the noise element sound data 231 of the first embodiment.

[0070] The FFT 533 performs fast Fourier transform on the audio signal S input from the remote control device 25 , and outputs Fourier coefficients, which are frequency domain information of the audio signal S, to the subtractor 534 .

[0071] The subtractor 534 subtracts the Fourier coefficient output from the frequency domain adaptive filter 532 from the Fourier coefficient input from the FFT 533 for each frequency, and outputs the result to the frequency domain adaptive filter 532 and the IFFT 535 .

[0072] The IFFT 535 performs an inverse fast Fourier transform on the Fourier coefficients outputted from the subtractor 534 , and outputs the Fourier coefficients to the speaker 24 as an output signal S′.

[0073] The frequency domain adaptive filter 532 performs filtering processing on the Fourier coefficients stored in the noise element sound data 531 , and outputs the result to the subtractor 534 .

[0074] In addition, the frequency domain adaptive filter 532 uses the Fourier coefficients output by the subtractor 534 as errors, and adapts the filter coefficients of the filtering process applied to the Fourier coefficients stored in the noise element sound data 531 through an adaptive algorithm such as LMS or NLMS to minimize the errors.

[0075] According to such a second embodiment, signal processing equivalent to the operation of the audio signal processing device 23 of the first embodiment can be performed in the frequency domain. As in the first embodiment, an audio signal obtained by reducing the level of only the approximate noise component from the audio signal S can be generated as an output signal S′ and output from the speaker 24.

[0076] Next, a third embodiment of the present invention will be described.

[0077] The third embodiment uses wavelet transform instead of Fourier transform to perform signal processing equivalent to the operation of the audio signal processing device 23 of the first embodiment, and only the configuration of the audio signal processing device 23 is different.

[0078] Figure 6 The configuration of an audio signal processing device 23 according to the third embodiment is shown.

[0079] As shown in the figure, the audio signal processing device 23 includes noise element sound data 631, an inverse wavelet transform unit 632 for performing inverse wavelet transform, an adaptive filter 633, and a subtractor 634. The subtractor 634 subtracts the output of the adaptive filter 633 from the audio signal S input from the remote control device 25, and outputs it to the speaker 24 as an output signal S'.

[0080] Wavelet transform coefficients are stored in the noise element sound data 631 , and the inverse wavelet transform unit 632 outputs a reference sound generated by performing inverse wavelet transform on the wavelet transform coefficients to the adaptive filter 633 .

[0081] The adaptive filter 633 applies filtering processing to the input reference sound and outputs it to the subtractor 634. In addition, the adaptive filter 633 uses the output signal S' output by the subtractor 634 as an error, and adapts the filter coefficients of the filtering processing applied to the reference sound through an adaptive algorithm such as LMS or NLMS to minimize the error.

[0082] The wavelet transform coefficients stored in the noise element sound data 631 are used to express a composite sound of pure tones of a plurality of frequencies extracted at predetermined frequency intervals from the removal target frequency band using information in the frequency domain, similarly to the first embodiment.

[0083] That is, the wavelet transform coefficients stored in the noise element sound data 631 are determined so that the reference sound after inverse wavelet transform becomes a composite sound of pure tones of multiple frequencies extracted at predetermined frequency intervals from the removal target frequency band as a predetermined frequency band.

[0084] Therefore, the noise element sound data 631 and the inverse wavelet transform unit 632 function as a sound source device that outputs the above-mentioned synthetic sound of the pure tone.

[0085] Here, the wavelet transform can obtain information in the frequency domain and the time domain, but in the third embodiment, the wavelet transform coefficients stored in the noise element sound data 631 do not show temporal changes, so the size of the pure sound components of each frequency included in the reference sound does not change over time.

[0086] The wavelet transform coefficients stored in the noise element sound data 631 are equivalent to T(a, b) obtained by, for example, the following equation 3 using the synthetic sound as f(t).

[0087] Here, a represents the scale parameter, b represents the translation parameter, and Ψ a，b represents a wavelet with scale a and translation b, and * represents the complex conjugate.

[0088] In addition, for example, the sound f(t) obtained by inverse wavelet transforming the wavelet transform coefficients T(a, b) stored in the noise element sound data 631 according to equation 4 becomes the reference sound. Ψ It is a constant determined by the type of mother wavelet.

[0089] [Formula 2]

[0090]

[0091] According to such a third embodiment, wavelet transform can be used instead of Fourier transform to perform signal processing equivalent to the operation of the audio signal processing device 23 of the first embodiment. As in the first embodiment, an audio signal obtained by reducing the level of only the approximate noise component from the audio signal S can be generated as an output signal S' and output from the speaker 24.

[0092] Next, a fourth embodiment of the present invention will be described.

[0093] The fourth embodiment performs signal processing equivalent to the operation of the audio signal processing device 23 of the third embodiment in the wavelet transform coefficient domain, and only the configuration of the audio signal processing device 23 is different.

[0094] Figure 7 The configuration of an audio signal processing device 23 according to the fourth embodiment is shown.

[0095] As shown in the figure, the audio signal processing device 23 according to the second embodiment includes noise element sound data 731 , a wavelet transform coefficient domain adaptive filter 732 , a wavelet transform unit 733 , a subtractor 734 , and an inverse wavelet transform unit 735 .

[0096] The noise element sound data 731 stores the same wavelet transform coefficients as those of the noise element sound data 631 of the third embodiment.

[0097] The wavelet transform unit 733 performs wavelet transform on the audio signal S input from the remote control device 25 , and outputs wavelet transform coefficients expressing the audio signal S using information in the frequency domain to the subtractor 734 .

[0098] The subtractor 734 subtracts the wavelet transform coefficient output by the wavelet transform coefficient domain adaptive filter 732 from the wavelet transform coefficient input from the wavelet transform unit 733 for each wavelet, and outputs the result to the wavelet transform coefficient domain adaptive filter 732 and the inverse wavelet transform unit 735 .

[0099] The inverse wavelet transform unit 735 performs inverse wavelet transform on the wavelet transform coefficients output from the subtractor 734 , and outputs the inverse wavelet transform to the speaker 24 as an output signal S′.

[0100] The wavelet transform coefficient domain adaptive filter 732 performs filtering processing on the wavelet transform coefficients stored in the noise element sound data 731 , and outputs the result to the subtractor 734 .

[0101] In addition, the wavelet transform coefficient domain adaptive filter 732 uses the wavelet transform coefficient output by the subtractor 734 as an error, and adapts the filter coefficients of the filtering process applied to the wavelet transform coefficients stored in the noise element sound data 731 through an adaptive algorithm such as LMS or NLMS to minimize the error.

[0102] According to such a fourth embodiment, signal processing equivalent to the operation of the audio signal processing device 23 of the third embodiment can be performed in the wavelet transform coefficient domain. As in the third embodiment, an audio signal obtained by reducing the level of only the approximate noise component from the audio signal S can be generated as an output signal S' and output from the speaker 24.

[0103] Next, a fifth embodiment of the present invention will be described.

[0104] The fifth embodiment performs signal processing equivalent to the operation of the audio signal processing device 23 of the first embodiment without using Fourier transform, and only the configuration of the audio signal processing device 23 is different.

[0105] Figure 8 The configuration of an audio signal processing device 23 according to the fifth embodiment is shown.

[0106] As shown in the figure, the audio signal processing device 23 includes a sound source device 831, an adaptive filter 832, and a subtractor 733. The sound source device 831 is set with audio data of a synthetic sound of pure sounds of multiple frequencies extracted at predetermined frequency intervals within a removal object frequency band. The subtractor 733 subtracts the output of the adaptive filter 832 from the audio signal S input from the remote control device 25, and outputs it to the speaker 24 as an output signal S'.

[0107] The sound source device 831 reproduces the set audio data and outputs it to the adaptive filter 832 as a reference sound.

[0108] The adaptive filter 832 applies filtering processing to the input reference sound and outputs it to the subtractor 833. In addition, the adaptive filter 832 uses the output signal S' output by the subtractor 833 as an error, and adapts the filter coefficients of the filtering processing applied to the reference sound through an adaptive algorithm such as LMS or NLMS to minimize the error.

[0109] According to such a fifth embodiment, it is possible to perform signal processing equivalent to the operation of the audio signal processing device 23 of the first embodiment without using Fourier transform. As in the third embodiment, it is possible to generate an audio signal obtained by reducing the level of only the approximate noise component from the audio signal S as an output signal S' and output it from the speaker 24.

[0110] The embodiments of the present invention have been described above.

[0111] As described above, according to each embodiment, with a simple configuration that does not require a plurality of microphones, it is possible to suppress unknown noise components included in the audio signal S and output the audio signal as the output signal S′.

[0112] Here, an example of application to a remote control system has been described above, but the above embodiments can be used in any system to reduce the level of the noise component in the sound obtained by mixing the listening target sound and noise picked up by the microphone 14 .

[0113] For example, if the above embodiments are applied to an online conference system, without relying on the noise elimination function of the mobile terminal, it is possible to reduce the noise level from the received sound received by participants participating in the online conference using mobile terminals in transportation, and output it to the speakers 24 of other participants' terminals.

Claims

1. An audio signal processing device that suppresses the noise component of an input signal including a listening object sound component and a noise component and outputs the signal as an output signal, characterized in that: have: A synthetic sound output mechanism outputs a synthetic sound signal; an adaptive filter, taking the synthesized sound signal output by the synthesized sound output mechanism as input; as well as a subtraction operation mechanism, which subtracts the output signal of the adaptive filter from the input signal and outputs the output signal as the output signal, The adaptive filter performs adaptive operation by taking the output signal output by the subtraction operation mechanism as an error. The synthetic sound output means outputs, as the synthetic sound signal, a signal of a synthetic sound of pure tones of a plurality of frequencies extracted from a removal target frequency band, which is a frequency band of a noise component to be suppressed in the audio signal processing device.

2. The audio signal processing device according to claim 1, characterized in that: The synthetic sound output unit outputs, as the synthetic sound signal, a signal of a synthetic sound of pure tones of a plurality of frequencies extracted at regular frequency intervals from within the removal target frequency band.

3. The audio signal processing device according to claim 1, characterized in that: The synthetic sound output mechanism comprises: an information storage mechanism for storing frequency domain information of the synthesized sound signal to be output; and The conversion mechanism converts the information stored in the information storage mechanism into the synthetic sound signal.

4. The audio signal processing device according to claim 3, characterized in that: The frequency domain information of the synthesized sound signal is Fourier coefficients. The conversion mechanism performs conversion into the synthetic sound signal by inverse fast Fourier transform.

5. The audio signal processing device according to claim 3, characterized in that: The frequency domain information of the synthetic sound signal is wavelet transform coefficients. The conversion means performs conversion into the synthetic audio signal by inverse wavelet transformation.

6. An audio signal processing device that suppresses the noise component of an input signal including a listening object sound component and a noise component and outputs the signal as an output signal, characterized in that: have: A first conversion mechanism that converts the input signal into first information that is information in the frequency domain of the input signal; The information storage means stores, as a synthetic sound signal, a signal of a synthetic sound of pure tones of a plurality of frequencies extracted from a frequency band to be removed, which is a frequency band of a noise component to be suppressed in the audio signal processing device, and frequency domain information of the synthetic sound signal as second information; an adaptive filter that takes the second information stored in the information storage means as input and outputs third information as frequency domain information of the noise-cancelled sound; a subtraction operation unit that outputs frequency domain information obtained by subtracting the third information from the first information as fourth information; as well as The second conversion means converts the fourth information into a sound signal and outputs the sound signal as the output signal. The adaptive filter performs an adaptive operation by treating the fourth information as an error.

7. The audio signal processing device according to claim 6, characterized in that: The synthetic audio signal is a synthetic sound signal of pure tones of a plurality of frequencies extracted at regular frequency intervals from the removal target frequency band.

8. The audio signal processing device according to claim 6, characterized in that: The information in the frequency domain is the Fourier coefficients, The first conversion mechanism performs a fast Fourier transform to convert the input signal into the first information. The second conversion means performs inverse fast Fourier transform to convert the fourth information into the output signal.

9. The audio signal processing device according to claim 6, characterized in that: The information in the frequency domain is the wavelet transform coefficients, The first conversion means performs wavelet transformation to convert the input signal into the first information. The second conversion means performs inverse wavelet transform to convert the fourth information into the output signal.

10. A remote control system comprising the audio signal processing device according to any one of claims 1 to 9, characterized in that: have: A vehicle-side system, mounted on a vehicle; and The control side system controls the vehicle remotely. The vehicle-side system comprises: a microphone that picks up a sound obtained by mixing the sound of the listening object and the noise and outputs the sound as a monitoring sound signal; and The sending mechanism sends the monitoring sound signal output by the microphone to the control side system, The control side system comprises: A receiving mechanism receives the monitoring sound signal from the vehicle-side system; Speakers; and The audio signal processing device, The monitoring audio signal received by the receiving means is input as the input signal to the audio signal processing device, and the output signal is output from the audio signal processing device to the speaker.

Images