Acoustic signal processing device

Abstract

This acoustic signal processing device (10) comprises: a first left signal generation unit (14LD) that generates a first left propagation signal corresponding to a sound from a left virtual sound source (3L) to a left ear (91L); a second left signal generation unit (14LX) that generates a second left propagation signal corresponding to a sound from the left virtual sound source (3L) to a right ear (91R); a first right signal generation unit (14RD) that generates a first right propagation signal corresponding to a sound from a right virtual sound source (3R) to the right ear (91R); and a second right signal generation unit (14RX) that generates a second right propagation signal corresponding to a sound from the right virtual sound source (3R) to the left ear (91L). These signal generation units each include a filter unit (21) that divides a corresponding input signal into a plurality of band signals of mutually different frequency bands, a delay processing unit (22) that performs delay processing on each of the plurality of band signals such that the delay time becomes longer as the frequency band is lower, and a remixing unit (23) that remixes the plurality of band signals on which the delay processing has been performed to generate a corresponding propagation signal.

Description

Acoustic signal processing device

[0001] The present invention relates to an acoustic signal processing device.

[0002] Patent Document 1 discloses an acoustic signal processing technology in which a direct sound signal corresponding to a direct sound from left and right virtual sound sources, a reflected sound signal corresponding to a reflected sound of the direct sound, and a crosstalk sound signal corresponding to a crosstalk sound in both ears of a listener are mixed in a mixer, and the output of the mixer is supplied to a headphone speaker. The crosstalk signal is formed by subjecting an input signal to band division processing and delay processing. The direct sound signal is formed by delaying the input signal. Taking into account the transmission time difference between the direct sound and the reflected sound, the reflected sound signal is formed by delaying the direct sound signal.

[0003] Japanese Patent Application Laid-Open No. 2004-23486

[0004] The above technology does not specifically consider delay processing for crosstalk sound signals. Furthermore, only delay processing for compensating for the delay of the crosstalk signal is performed on the direct sound signal. Therefore, it is difficult to make the listener feel as if the virtual sound source is located far away from the listener. It is also difficult to adjust the sense of distance between the virtual sound source and the listener.

[0005] An object of the present invention is to provide an acoustic signal processing device that can easily adjust the perceived distance between a virtual sound source and a listener.

[0006] One aspect of the present invention provides an acoustic signal processing device comprising: a first signal generation unit that processes an input signal and generates a first propagation signal corresponding to sound propagating from a virtual sound source to an ear that is closer to the virtual sound source; and a second signal generation unit that processes the input signal and generates a second propagation signal corresponding to sound propagating from the virtual sound source to an ear that is farther from the virtual sound source, wherein the first signal generation unit and the second signal generation unit each include a filter unit that separates the corresponding input signal into a plurality of band signals having mutually different frequency bands; a delay processing unit that delays each of the plurality of band signals so that the delay time becomes longer as the frequency band becomes lower; and a remixing unit that mixes the delayed plurality of band signals to generate a corresponding propagation signal.

[0007] According to the above configuration, when converting an input signal into left and right output signals, two types of propagation signals corresponding to sounds propagated from a virtual sound source to the left and right ears are generated. Each propagation signal is generated by dividing the corresponding input signal into multiple band signals, delaying the multiple band signals, and remixing the delayed multiple band signals. In this delay process, the multiple band signals are each processed such that the delay time is longer for lower frequency bands.

[0008] This allows distance information to be added to the left and right output signals. By adjusting the delay time for each band signal, it becomes possible to adjust the sense of distance that the listener perceives between themselves and the virtual sound source.

[0009] At least one of the first signal generation unit and the second signal generation unit may include a direct signal generation unit that generates a direct propagation signal corresponding to sound that propagates directly from a corresponding virtual sound source to a corresponding ear, and a reflected signal generation unit that generates a reflected propagation signal corresponding to sound that is reflected from the corresponding virtual sound source and propagates to the corresponding ear.

[0010] According to the above configuration, reflections can be taken into consideration, and the listener can perceive a clearer sound image.

[0011] According to the present invention, it is possible to provide an acoustic signal processing device that can easily adjust the perceived distance between a virtual sound source and a listener.

[0012] A conceptual diagram showing an audio reproduction device including an audio signal processing device according to a first embodiment. A schematic configuration diagram of the audio signal processing device according to the first embodiment. A conceptual diagram showing an audio reproduction device including an audio signal processing device according to a second embodiment. A schematic configuration diagram of a portion of the audio signal processing device according to the second embodiment. A schematic configuration diagram of a portion of the audio signal processing device according to the second embodiment. A graph showing an output signal of the audio signal processing device according to the present invention.

[0013] Hereinafter, embodiments will be described with reference to the drawings. Note that the same or corresponding elements are designated by the same reference numerals throughout the drawings, and detailed descriptions thereof will be omitted.

[0014] 1 shows a stereo earphone as an example of an audio reproduction device 1. The audio reproduction device 1 includes an acoustic signal processing device 10 (see FIG. 2) according to this embodiment and a stereo audio output unit. The audio output unit includes a left speaker 2L and a right speaker 2R.

[0015] 1 and 2 , the audio signal processing device 10 is realized, for example, by an electric circuit mounted on a board housed in the audio reproduction device 1. The audio signal processing device 10 has a left input terminal 11L and a right input terminal 11R to which left and right channel stereo audio signals are input as left input signals and right input signals. The audio signal processing device 10 converts the left input signal and right input signal into a left output signal and a right output signal. The audio signal processing device 10 has a left output terminal 12L and a right output terminal 12R to which left output signals and right output signals are output as left and right channel stereo audio signals.

[0016] The audio output unit is connected to output terminals 12L and 12R. The left speaker 2L is connected to the left output terminal 12L and outputs a left output signal as audio. The right speaker 2R is connected to the right output terminal 12R and outputs a right output signal as audio.

[0017] In this embodiment, the audio reproduction device 1 is an earphone or a headphone, and the left speaker 2L is worn on the left ear 91L of the listener 90 and outputs audio into the outer ear of the left ear 91L. The right speaker 2R is worn on the right ear 91R of the listener 90 and outputs audio into the outer ear of the right ear 91R.

[0018] Referring to FIG. 1, in the signal conversion process of the audio reproduction device 1, a left virtual sound source 3L and a right virtual sound source 3R, and a sound propagation path 4 from the virtual sound sources 3L and 3R to ears 91L and 91R of a listener 90 are virtually assumed.

[0019] The positions of the virtual sound sources 3L and 3R can be adjusted in the front-to-rear direction of the listener 90, and are localized, for example, at a position away from the head 92 of the listener 90 in front of the listener. The propagation paths 4 include four systems: a first left propagation path 4LD, a second left propagation path 4LX, a first right propagation path 4RD, and a second right propagation path 4RX. The first left propagation path 4LD is a path from the left virtual sound source 3L to the left ear 91L. The second left propagation path 4LX is a path from the left virtual sound source 3L to the right ear 91R. The first right propagation path 4RD is a path from the right virtual sound source 3R to the right ear 91R. The second right propagation path 4RX is a path from the right virtual sound source 3R to the left ear 91L. In this embodiment, all of the propagation paths 4 are linear. The first left propagation path 4LD and the first right propagation path 4RD are paths between the corresponding virtual sound source and the ear closest to the virtual sound source. The second left propagation path 4LX and the second right propagation path 4RX are paths between the corresponding virtual sound source and the ear farther from the virtual sound source.

[0020] The listener 90 may perceive the sound as if it is being output from left and right virtual sound sources 3L and 3R located far in front of him, even though the sound is actually being output from the sound output unit to his outer ears. The listener 90 may perceive the sound as if it is reaching his left ear 91L from the virtual sound sources 3L and 3R via two propagation paths 4LD and 4RX, and as reaching his right ear 91R via two propagation paths 4RD and 4LX.

[0021] Next, the configuration of an audio signal processing device 10 that realizes this sensation will be described with reference to Fig. 2. Note that the "signals" processed between the input terminals 11L, 11R and the output terminals 12L, 12R are electrical audio signals.

[0022] In addition to the left input terminal 11L, right input terminal 11R, left output terminal 12L, and right output terminal 12R described above, the acoustic signal processing device 10 also includes a left input amplifier 13L, a right input amplifier 13R, a first left signal generator 14LD, a second left signal generator 14LX, a first right signal generator 14RD, a second right signal generator 14RX, a left mixer 15L, a right mixer 15R, a left output amplifier 16L, and a right output amplifier 16R.

[0023] The left input amplifier 13L amplifies the left input signal input to the left input terminal 11L. The right input amplifier 13R amplifies the right input signal input to the right input terminal 11R. The left input signal corresponds to the left virtual sound source 3L. The right input signal corresponds to the right virtual sound source 3R.

[0024] The first left signal generator 14LD processes the left input signal amplified by the left input amplifier 13L to generate a first left propagation signal. The first left propagation signal is a signal corresponding to sound propagating from the left virtual sound source 3L to the left ear 91L via the first left propagation path 4LD.

[0025] The second left signal generator 14LX processes the left input signal amplified by the left input amplifier 13L to generate a second left propagation signal. The second left propagation signal is a signal corresponding to sound propagating from the left virtual sound source 3L to the right ear 91R via the second left propagation path 4LX.

[0026] The first right signal generator 14RD processes the right input signal amplified by the right input amplifier 13R to generate a first right propagation signal. The first right propagation signal is a signal corresponding to sound propagating from the right virtual sound source 3R to the right ear 91R via the first right propagation path 4RD.

[0027] The second right signal generator 14RX processes the right input signal amplified by the right input amplifier 13R to generate a second right propagation signal. The second right propagation signal is a signal corresponding to sound propagating from the right virtual sound source 3R to the left ear 91L via the second right propagation path 4RX.

[0028] The left mixer 15L mixes the first left propagation signal and the second right propagation signal to generate a left output signal. The left output amplifier 16L amplifies the left output signal generated by the left mixer 15L. The left output terminal 12L outputs the left output signal amplified by the left output amplifier 16L.

[0029] The right mixer 15R mixes the first right propagation signal and the second left propagation signal to generate a right output signal. The right output amplifier 16R amplifies the right output signal generated by the right mixer 15R. The right output terminal 12R outputs the right output signal amplified by the right output amplifier 16R.

[0030] With the left speaker 2L attached to the left ear 91L and the right speaker 2R attached to the right ear 91R, the listener 90 can perceive as if the sound output from the speakers 2L and 2R is propagating from the left virtual sound source 3L to the left ear 91L and right ear 91R via the left propagation paths 4LD and 4LX. The listener 90 can also perceive as if the sound output from the speakers 2L and 2R is propagating from the right virtual sound source 3R to the left ear 91L and right ear 91R via the right propagation paths 4RD and 4RX.

[0031] The first left signal generating unit 14LD, the second left signal generating unit 14LX, the first right signal generating unit 14RD, and the second right signal generating unit 14RX each include a filter unit 21, a delay processing unit 22, a remixing unit 23, and a level adjustment unit 24.

[0032] The filter unit 21 divides the corresponding input signal into a plurality of band signals having different frequency bands. The filter unit 21 includes a first left filter unit 21LD, a second left filter unit 21LX, a first right filter unit 21RD, and a second right filter unit 21RX corresponding to the four signal generation units 14LD, 14LX, 14RD, and 14RX, respectively.

[0033] With respect to the first left signal generation unit 14LD and the second left signal generation unit 14LX that generate the propagation signal from the left virtual sound source 3L, the "corresponding input signal" is the left input signal. In this embodiment, the filter unit 21 is common to these two signal generation units 14LD, 14LX. That is, the first left filter unit 21LD and the second left filter unit 21LX are unified as the left filter unit 21L. The left filter unit 21L divides the left input signal into a plurality of band signals to generate the propagation signals (first left propagation signal and second left propagation signal) from the left virtual sound source 3L.

[0034] With respect to the first right signal generation unit 14RD and the second right signal generation unit 14RX that generate the propagation signals from the right virtual sound source 3R, the "corresponding input signal" is the right input signal. In this embodiment, the filter unit 21 is common to these two signal generation units 14RD, 14RX. That is, the first right filter unit 21RD and the second right filter unit 21RX are unified as the right filter unit 21R. The right filter unit 21R divides the right input signal into a plurality of band signals to generate the propagation signals (first right propagation signal and second right propagation signal) from the right virtual sound source 3R.

[0035] In this embodiment, as a mere example, each of the left filter unit 21L and the right filter unit 21R obtains three band signals: a high-band signal, a mid-band signal, and a low-band signal. The high-band signal is a signal obtained by extracting high-frequency components (e.g., 2 kHz or higher) from the corresponding input signal. The mid-band signal is a signal obtained by extracting mid-frequency components (e.g., 1 kHz to 2 kHz) from the corresponding input signal. The low-band signal is a signal obtained by extracting low-frequency components (e.g., 1 kHz or lower) from the corresponding input signal. To achieve this, each of the left filter unit 21L and the right filter unit 21R may have, for example, a high-pass filter for obtaining the high-band signal, a low-pass filter for obtaining the low-band signal, and a band-pass filter for obtaining the mid-band signal.

[0036] The delay processing unit 22 includes a first left delay processing unit 22LD, a second left delay processing unit 22LX, a first right delay processing unit 22RD, and a second right delay processing unit 22RX corresponding to the four signal generating units 14LD, 14LX, 14RD, and 14RX, respectively. The first left delay processing unit 22LD and the second left delay processing unit 22LX receive multiple (e.g., three) band signals obtained by dividing the left input signal from the left filter unit 21L. The first right delay processing unit 22RD and the second right delay processing unit 22RX receive multiple (e.g., three) band signals obtained by dividing the right input signal from the right filter unit 21R.

[0037] The delay processing unit 22 performs two types of delay processing on the input multiple band signals.

[0038] First, delay processing is performed according to the frequency band of the band signal to add distance information of the propagation path from the virtual sound source to the ear. In this type of delay processing, different band delay times are set for the multiple band signals in each delay processing unit 22. Each delay processing unit 22 delays the multiple band signals such that the band delay time is longer for lower frequency bands. The band delay time for low band signals is called the "low-band delay time," the band delay time for mid-band signals is called the "mid-band delay time," and the band delay time for high band signals is called the "high-band delay time."

[0039] The low-band signal is delayed by a low-band delay time. The mid-band signal is delayed by a mid-band delay time. The high-band signal is delayed by a high-band delay time. The low-band delay time is longer than the mid-band delay time, and the mid-band delay time is longer than the high-band delay time. The low-band delay time is, for example, 2 milliseconds to 4 milliseconds, the mid-band delay time is, for example, 1 millisecond to 2 milliseconds, and the high-band delay time is, for example, 0 to 0.5 milliseconds. The low-band delay time, mid-band delay time, and high-band delay time may be the same or different among the four delay processing units 22LD, 22LX, 22RD, and 22RX.

[0040] Second, delay processing is performed according to the corresponding propagation path difference. For example, the second left propagation path 4LX from the left virtual sound source 3L to the right ear 91R is longer than the first left propagation path 4LD to the left ear 91L. Therefore, the time when the sound from the left virtual sound source 3L reaches the right ear 91R is delayed by ΔtL (e.g., 0 to 1 millisecond) relative to the time when the sound reaches the left ear 91L. The second left delay processing unit 22LX delays the input multiple band signals according to this propagation time difference ΔtL. In this type of delay processing, a uniform propagation delay time may be set among the multiple band signals in each delay processing unit 22. In the first left delay processing unit 22LD, the propagation path difference is 0, so the propagation delay time is 0 milliseconds.

[0041] The same applies to the first right delay processing unit 22RD and the second right delay processing unit 22RX. Since the second right propagation path 4RX from the right virtual sound source 3R to the left ear 91R is longer than the first right propagation path 4RD to the right ear 91R, the time when the sound from the right virtual sound source 3R reaches the left ear 91L is delayed by ΔtR (for example, 0 to 1 millisecond) relative to the time when the sound reaches the right ear 91R.

[0042] As a result of the above, each band signal in each delay processing unit 22 is delayed by a delay time corresponding to the sum of the propagation delay time corresponding to the delay processing unit 22 and the band delay time corresponding to the frequency band of the band signal.

[0043] Note that for a band signal with the highest frequency band (in this example, a high-band signal), its band delay time (in this example, a high-band delay time) may be zero. In other words, the high-band signal may not be subject to delay processing according to the frequency band. Even in this case, by setting a low-band delay time and a mid-band delay time, the delay is longest for the band signal with the lowest frequency band (low-band signal). Therefore, the lower the frequency band, the longer the delay time. Furthermore, since a propagation delay time can also be set for the high-band signal, a positive delay time can be set for the high-band signal even if the high-band delay time is zero.

[0044] The remixing unit 23 remixes the delayed band signals to generate corresponding propagation signals. The remixing unit 23 includes a first left remixing unit 23LD, a second left remixing unit 23LX, a first right remixing unit 23RD, and a second right remixing unit 23RX corresponding to the four signal generating units 14LD, 14LX, 14RD, and 14RX, respectively.

[0045] The first left remixer 23LD corresponds to the first left delay processor 22LD and the first left propagation signal. The first left remixer 23LD remixes the band signals delayed by the first left delay processor 22LD to generate a first left propagation signal and outputs the first left propagation signal to the left mixer 15L.

[0046] The second left recombining unit 23LX, the first right recombining unit 23RD, and the second right recombining unit 23RX are configured in the same manner. The second left recombining unit 23LX outputs a second left propagation signal to the right recombining unit 15R. The first right recombining unit 23RD outputs a first right propagation signal to the right recombining unit 15R. The second right recombining unit 23RX outputs a second right propagation signal to the left recombining unit 15L.

[0047] Humans perceive sound by collecting sound in the outer ear and transmitting it to the eardrum, amplifying the eardrum's vibrations in the middle ear, converting the vibrations into neural excitation (electrical signals) in the cochlea, and transmitting the neural excitation to the brain via the vestibulocochlear nerve. As is well known in the von Bekesy model, pure tone vibrations transmitted to the fluid at the base of the cochlea create fluid flow, vibrating the basilar membrane and traveling as waves toward the apex of the cochlea. These vibrations only travel a certain distance depending on the frequency. High-pitched sounds (sounds in the high-frequency range) transmit only a small amount of vibration, while low-pitched sounds (sounds in the low-frequency range) transmit all the way to the apex. The vibration of the basilar membrane is strongest just before this limit, and pure tones of different pitches each produce vibration patterns at different positions on the basilar membrane. This demonstrates how acoustic signals are transmitted to the brain at different frequencies, as if through a process similar to that of a fast Fourier transform (FFT).

[0048] The inventors discovered that when a sound containing multiple spectra is generated, the higher the frequency, the greater the phase advance of the acoustic signal, and that this phase advance is proportional to the distance. They also discovered that the phase shift according to frequency appears as a difference in the timing of the generation of the electrical signal output from the cochlea, and that the brain recognizes this time difference as a difference in distance.

[0049] From this, we came up with the idea that if we divided the input signal into frequency bands, delayed each of the multiple band signals so that the delay time becomes longer the lower the frequency band, and then remixed the delayed multiple band signals, we could create a sense of distance between the virtual sound source and the listener.We then came up with the idea that we could adjust the distance between the virtual sound source and the listener by adjusting the band delay time added to the multiple band signals.Specifically, we came up with the idea that a long delay time in the low range would make the sound source seem farther away, and a short delay time would make the sound source seem closer.

[0050] In the acoustic signal processing device 10 according to this embodiment, the first left propagation signal and the second right propagation signal generated as described above are mixed in the left mixer 15L to generate a left output signal. The first right propagation signal and the second left propagation signal generated as described above are mixed in the right mixer 15R to generate a right output signal. This adds distance information to each propagation signal, and ultimately adds distance information to the left and right output signals. This makes it possible to easily adjust the sense of distance between the virtual sound sources 3L and 3R and the listener 90.

[0051] The filter section 21 is common to the first left signal generation section 14LD and the second left signal generation section 14LX, and is common to the first right signal generation section 14RD and the second right signal generation section 14RX, thereby simplifying the configuration of the acoustic signal processing device 10.

[0052] The level adjustment unit 24 adjusts the levels of the band signals before remixing. The level adjustment unit 24 adjusts the levels depending on the distance and direction of the corresponding propagation path. For example, because the first left propagation path 4LD is shorter than the second left propagation path 4LX, the level adjustment unit 24 adjusts the levels of each band signal of the first left propagation signal so that they are higher than the levels of each band signal of the second left propagation signal. The same applies to the first right propagation signal and the second right propagation signal. The level adjustment unit 24 also adjusts the levels of each band signal of the propagation signals corresponding to crosstalk sound (the second left propagation signal and the second right propagation signal) depending on the frequency. It is said that sound arriving from a sound source diagonally in front (a virtual sound source on the right side for the left ear, and a virtual sound source on the left side for the right ear) becomes more difficult to hear the higher the frequency. Taking this into consideration, the level adjustment unit 24 adjusts the levels of each band signal of the second left propagation signal so that they are lower the higher the frequency. The same applies to the second right propagation signal. Such level adjustment allows the listener to perceive a clearer sound image.

[0053] Next, the acoustic signal processing device 10 according to the second embodiment will be described with reference to FIGS. 3 to 5, focusing on the differences from the first embodiment.

[0054] Referring to Figure 3, in the signal conversion processing of the audio reproduction device 1 according to this embodiment, a first left propagation path 4LD, a second left propagation path 4LX, a first right propagation path 4RD, and a second right propagation path 4RX are hypothetically assumed as sound propagation paths 4 from the virtual sound sources 3L, 3R to the ears 91L, 91R of the listener 90.

[0055] The first left propagation path 4LD includes a first left direct path 4LDd that runs directly from the left virtual sound source 3L to the left ear 91L, and a first left reflected path 4LDr that runs from the left virtual sound source 3L to the left ear 91L via the shoulder 93. The other three propagation paths are similar. The second left propagation path 4LX includes a second left direct path 4LXd and a second left reflected path 4LXr. The first right propagation path 4RD includes a first right direct path 4RDd and a first right reflected path 4RDr. The second right propagation path 4RX includes a second right direct path 4RXd and a second right reflected path 4RXr.

[0056] The direct paths 4LDd, 4LXd, 4RDd, and 4RXd are linear and correspond to the first left propagation path 4LD, the second left propagation path 4LX, the first right propagation path 4RD, and the second right propagation path 4RX in the first embodiment (see FIG. 1). The reflected paths 4LDr, 4LXr, 4RDr, and 4RXr are polygonal.

[0057] 4, the first left signal generator 14LD includes a first left direct signal generator 14LDd and a first left reflected signal generator 14LDr. The first left direct signal generator 14LDd processes the left input signal amplified by the left input amplifier 13L to generate a first left direct propagation signal corresponding to sound propagating directly from the left virtual sound source 3L to the left ear 91L via the first left direct path 4LDd. The first left reflected signal generator 14LDr processes the left input signal amplified by the left input amplifier 13L to generate a first left reflected propagation signal corresponding to sound propagating from the left virtual sound source 3L to the left ear 91L after being reflected by the shoulder 93 via the first left reflection path 4LDr.

[0058] The first left direct signal generator 14LDd and the first left reflected signal generator 14LDr share the filter unit 21. The first left direct signal generator 14LDd and the first left reflected signal generator 14LDr each have a delay processing unit 22 and a remixing unit 23.

[0059] In both the first left direct signal generation unit 14LDd and the first left reflected signal generation unit 14LDr, the delay processing unit 22 performs delay processing on each of the multiple band signals separated by the filter unit 21. Because the first left reflected path 4LDr is longer, the propagation delay time set by the first left reflected signal generation unit 14LDr is longer than the propagation delay time set by the first left direct signal generation unit 14LDd. As in the first embodiment, the delay processing unit 22 also performs delay processing according to the frequency band. Furthermore, taking into account the difference in distance between the first left direct path 4LDd and the first left reflected path 4LDr, the level adjustment unit 24 adjusts the level of the propagation signal so that the level of the first left reflected propagation signal is lower than the level of the first left direct propagation signal. The first left direct signal generation unit 14LDd generates the first left direct propagation signal by mixing the multiple band signals in the remixing unit 23. The first left reflected signal generation unit 14LDr generates the first left reflected propagation signal by mixing the multiple band signals in the remixing unit 23. The first left direct propagation signal and the first left reflected propagation signal are mixed in the left mixer 15L.

[0060] 4 and 5 , in this embodiment, the second left signal generator 14LX, the first right signal generator 14RD, and the second right signal generator 14RX are also similar. The second left signal generator 14LX includes a second left direct signal generator 14LXd that generates a second left direct propagation signal and a second left reflected signal generator 14LXr that generates a second left reflected propagation signal. The first right signal generator 14RD includes a first right direct signal generator 14RDd that generates a first right direct propagation signal and a first right reflected signal generator 14RDr that generates a first right reflected propagation signal. The second right signal generator 14RX includes a second right direct signal generator 14RXd that generates a second right direct propagation signal and a second right reflected signal generator 14RXr that generates a second right reflected propagation signal.

[0061] The second right direct signal and the second right reflected signal are mixed together with the first left direct signal and the first left reflected signal in the left mixer 15L. The second left direct signal, the second left reflected signal, the first right direct signal, and the first right reflected signal are mixed in the right mixer 15R.

[0062] According to this embodiment, shoulder reflex can be taken into consideration, allowing the listener to perceive a clearer sound image. Since shoulder reflex is taken into consideration in all of the first left signal generator 14LD, the second left signal generator 14LX, the first right signal generator 14RD, and the second right signal generator 14RX, the sound image can be made clearer.

[0063] Although the embodiments have been described above, the above configurations can be modified as appropriate within the scope of the present invention.

[0064] For example, the delay processing unit 22 may divide the corresponding input signal into a number of band signals other than three (for example, two or four or more).

[0065] In the second embodiment, the first left signal generating unit 14LD, the second left signal generating unit 14LX, the first right signal generating unit 14RD, and the second right signal generating unit 14RX all include a direct signal generating unit and a reflected signal generating unit. This is merely an example, and at least one of the first left signal generating unit 14LD, the second left signal generating unit 14LX, the first right signal generating unit 14RD, and the second right signal generating unit 14RX may include a direct signal generating unit and a reflected signal generating unit.

[0066] In the second embodiment, reflection from the shoulder 93 is assumed. This is merely an example, and the reflected propagation signal and the reflected path may be assumed to be reflected from other reflectors such as a wall. Furthermore, multiple reflectors may be assumed. In other words, multiple reflection paths may be assumed between the same virtual sound source and the same ear.

[0067] In the above embodiment and drawings, the level adjustment unit 24 is provided between the delay processing unit 22 and the recombining unit 23. This is merely an example. The level adjustment unit 24 may be located anywhere as long as it can adjust the level of each band signal. The level adjustment unit 24 may be located between the filter unit 21 and the delay processing unit 22, in the delay processing unit 22, or in the recombining unit 23.

[0068] In the above embodiment, there are two virtual sound sources, one on the left and one on the right. This is merely an example, and the number of virtual sound sources in the acoustic signal processing device 10 may be one, or three or more. In other words, the number of pairs consisting of "a left signal generation unit that processes an input signal and generates a left propagation signal corresponding to sound propagating from a virtual sound source to the left ear, and a right signal generation unit that processes the same input signal and generates a right propagation signal corresponding to sound propagating from the same virtual sound source to the right ear" may be one, two, three or more.

[0069] The acoustic signal processing device 10 can also be applied to an audio reproduction device 1 other than earphones, such as headphones or speakers installed in a room or the like.

[0070] In the above embodiment, the acoustic signal processing device 10 is realized as a circuit on a board housed in (the housing of) the audio reproduction device 1. This is merely one example, and the acoustic signal processing device 10 may be implemented as software in an application installed in an information terminal device (e.g., a personal computer, a smartphone, etc.). In this case, the acoustic signal processing device 10 processes the acoustic signal in the application in the same manner as described above. The output signal is input to an audio reproduction device connected to the information terminal device by wire or wirelessly.

[0071] FIG. 6 shows the waveform of an output signal from the audio signal processing device according to the present invention. As a mere example, the input signal is composed of a high-frequency sine wave signal and a low-frequency sine wave signal, with a predetermined interval τ (e.g., 10 milliseconds) between the high-frequency and low-frequency signals. "Output 1," "Output 2," ... "Output n" respectively represent the waveforms of the output signals from the remixing unit 23. The larger the k in the output k (Output 1, Output 2, ... Output n), the longer the delay time dk (d1, d2, ... dn) of the output relative to the high-frequency sine wave input, and the longer the delay time Δtk (Δt1, Δt2, ... Δtk) of the output relative to the low-frequency sine wave input. The output signal is obtained by combining these outputs 1 to n. The output signal is generated by superimposing sine wave outputs. The waveform is different from the impulse response.

[0072] REFERENCE SIGNS LIST 1 Audio reproduction device 2L, 2R Speakers 3L, 3R Virtual sound source 4 Propagation path 4LD First left propagation path 4LDd First left direct path 4LDr First left reflected path 4LX Second left propagation path 4LXd Second left direct path 4LXr Second left reflected path 4RD First right propagation path 4RDd First right direct path 4RDr First right reflected path 4RX Second right propagation path 4RXd Second right direct path 4RXr Second right reflected path 10 Acoustic signal processing device 11L, 11R Input terminal 12L, 12R Output terminal 13L, 13R Input amplifier 14LD First left signal generator 14LDd First left direct signal generator 14LDr First left reflected signal generator 14LX Second left signal generator 14LXd Second left direct signal generator 14LXr Second left reflected signal generation section 14RD First right signal generation section 14RDd First right direct signal generation section 14RDr First right reflected signal generation section 14RX Second right signal generation section 14RXd Second right direct signal generation section 14RXr Second right reflected signal generation section 15L Left mixing section 15R Right mixing section 16L, 16R Output amplifier section 21 Filter section 21L Left filter section 21LD First left filter section 21LX Second left filter section 21R Right filter section 21RD First right filter section 21RX Second right filter section 22 Delay processing section 22LD First left delay processing section 22LX Second left delay processing section 22RD First right delay processing section 22RX Second right delay processing section 23 Re-mixing section 23LD First left re-mixing section 23LX 2nd left remixing section 23RD 1st right remixing section 23RX 2nd right remixing section 24 Level adjustment section 90 Listener 91L Left ear 91R Right ear 92 Head 93 Shoulder

Claims

1. An acoustic signal processing device comprising: a first signal generation unit that processes an input signal and generates a first propagation signal corresponding to sound propagating from a virtual sound source to an ear closer to the virtual sound source; and a second signal generation unit that processes the input signal and generates a second propagation signal corresponding to sound propagating from the virtual sound source to an ear farther from the virtual sound source, wherein the first signal generation unit and the second signal generation unit each include: a filter unit that separates the corresponding input signal into a plurality of band signals having mutually different frequency bands; a delay processing unit that delays each of the plurality of band signals so that the delay time becomes longer as the frequency band becomes lower; and a remixing unit that remixes the delayed plurality of band signals to generate a corresponding propagation signal.

2. The acoustic signal processing device according to claim 1, wherein at least one of the first signal generating unit and the second signal generating unit includes: a direct signal generating unit that generates a direct propagation signal corresponding to sound that propagates directly from a corresponding virtual sound source to a corresponding ear; and a reflected signal generating unit that generates a reflected propagation signal corresponding to sound that is reflected from the corresponding virtual sound source and propagates to the corresponding ear.

Images