JPH0795698

Patent Translate
Powered by EPO and Google
Notice
This translation is machine-generated. It cannot be guaranteed that it is intelligible, accurate,
complete, reliable or fit for specific purposes. Critical decisions, such as commercially relevant or
financial decisions, should not be based on machine-translation output.
DESCRIPTION JPH0795698
[0001]
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an
audio reproducing apparatus suitable for use, for example, in reproducing an acoustic signal by
means of headphones.
[0002]
2. Description of the Related Art Heretofore, there has been a method of reproducing an acoustic
signal by means of headphones in which headphones are mounted on the head so as to cover
both ears of a listener and the acoustic signals are heard from both ears. In this method of
reproducing an acoustic signal by headphones, even if the signal from the signal source is a
stereo signal, a so-called in-head localization phenomenon occurs in which the sound image to be
reproduced is embedded in the head of the listener.
[0003]
On the other hand, there is a binaural sound pickup reproduction method as one of reproduction
methods of sound signals by headphones. The binaural sound pickup reproduction method refers
to the following method. A microphone called a dummy head microphone is provided in the holes
of the left and right ears of the dummy head assuming the head of the listener. The dummy head
microphone picks up an acoustic signal from the signal source. When the listener actually wears
10-05-2019
1
the headphones and reproduces the sound signal collected in this manner, a sense of realism is
obtained as if the sound from the signal source is being heard as it is. According to such a
binaural sound collecting and reproducing method, it is possible to improve the sense of
direction, the feeling of localization, the feeling of presence, etc. of the sound collection
reproduction sound image. However, in order to perform such binaural reproduction, a signal
source as a special source different from that for speaker reproduction, which is collected by a
dummy head microphone as a sound source signal, has been required.
[0004]
Therefore, it has been considered to obtain a reproduction effect in which a general stereo signal
is localized outside the head (speaker position) similar to speaker reproduction, for example, with
headphones by applying the above-described binaural sound pickup reproduction system.
However, in the case of stereo reproduction with a speaker, even if the listener changes the head
(face) direction, the absolute direction and position of the sound image does not change, and the
relative direction and position of the sound image felt by the listener changes Do. On the other
hand, in the case of binaural reproduction with headphones, even if the listener changes the head
(face) direction, the relative direction and position of the sound image perceived by the listener
does not change. For this reason, even in the case of binaural reproduction, when the listener
changes the head (face) direction, the sound field is formed in the listener's head, and in
particular, the sound image is localized in front of the listener. It was difficult to let Moreover, in
this case, the sound image tends to rise above the head.
[0005]
On the other hand, according to the headphone reproduction method described in Japanese
Patent Publication No. 42-227, the following binaural reproduction method using headphones is
considered. That is, since the sense of direction and the sense of localization of the sound image
are determined by the volume difference, the time difference, the phase difference, etc. of the
sounds heard by the left ear and the right ear, the system of the above-mentioned publication
uses audio signal lines of the left and right channels. A level control circuit and a variable delay
circuit are respectively provided, the head direction of the listener is detected, and the level
control circuit of the audio signal of each channel and the variable delay circuit are controlled by
the detection signal.
[0006]
10-05-2019
2
However, in the headphone reproduction method described in JP-B-42-227, the motor is driven
by the detection signal of the head direction of the listener itself, and this motor controls the
level control circuit and the variable resistor and variable capacitor of the variable delay circuit.
Since the listener is mechanically controlled by the analog signal, it causes a time delay until the
listener changes the head direction and changes the volume difference and time difference of the
audio signal of each channel supplied to the headphones, I could not cope with the movement of
the listener's head.
[0007]
In the headphone reproduction method described in JP-B-42-227, when the volume difference
and the time difference are changed, the change characteristic is the relative positional
relationship between the sound source and the listener, and the shape of the head of the listener.
And the shape of the pinna and so on.
That is, in the case of one change characteristic, the positional relationship between the sound
source and the listener is fixed, and the sense of distance and the distance between the sound
sources can not be changed. Because the shape is different, the degree of the effect may vary.
[0008]
Furthermore, according to the stereoscopic reproduction method described in Japanese Patent
Publication No. 54-19242, the relationship between the volume difference of the audio signal of
each channel supplied to the head of the listener and the headphones and the mutual variation
between time differences is continuously determined. It is stated that it is possible.
[0009]
However, in the three-dimensional reproduction system described in the above-mentioned
Japanese Patent Publication No. 54-19242, a large capacity memory is required to continuously
obtain the relationship between the volume difference of the audio signal and the mutual
variation of the time difference and store it. It was extremely difficult to achieve this.
[0010]
Furthermore, in the audio reproduction apparatus described in Japanese Patent Application LaidOpen No. 01-112900 filed by the same applicant as the present invention, the data of the volume
difference of these audio signals and the mutual relationship of the variation between them are
10-05-2019
3
not continuous but discretely data. An apparatus for processing audio signals is described.
[0011]
However, in the audio reproduction apparatus described in the above-mentioned JP-A-01112900, only the principle concept that can be applied to either analog or digital signal
processing is shown, and analog or digital signal processing is used, It lacks concreteness when
applied to actual products.
[0012]
The above-described conventional headphone reproduction method, stereo reproduction method
and audio reproduction apparatus described above each process these audio signals at a high
speed in response to the head movement of the listener. Although it can not be realized without
digital signal processing, its processing means and processing method are not shown, and it is
difficult to implement.
[0013]
In addition, in order to store the relationship between the volume difference of these audio
signals and the amount of change between time differences, it is necessary to provide a large
capacity of memory, which can also be realized without digital signal processing. Nevertheless,
the processing means and processing method are not shown, and there is a disadvantage that the
implementation is extremely difficult.
The present invention has been made in view of the above circumstances, and an object of the
present invention is to provide an audio reproduction device capable of high-speed processing in
accordance with the movement of the head of the listener while saving the memory capacity.
[0014]
SUMMARY OF THE INVENTION An audio reproduction apparatus according to the present
invention, as shown in FIG. 1, is disposed near a digital signal source 1 for supplying digital audio
signals of a plurality of channels, and in the vicinity of listeners' ears 23. Sound reproducing
means 24 for converting digital sound signal from digital signal source 1 into analog sound
signal by digital / analog converting means and reproducing it, and angle detection for detecting
head movement of listener 23 with respect to the reference direction at every predetermined
angle Means 28, address signal conversion means 34 for converting the angle detected by the
10-05-2019
4
angle detection means 28 into a digital address signal, and virtual sound source positions relative
to the reference direction of the head of the listener 23 for each angle that the listener 23 can
recognize From the digital signal source 1 and storage means 35 which digitally records the
impulse response from the listener 23 to both ears of the listener 23 Integral means 5, 6, 7, 8, 9,
10, 11, 12 performing convolutional integration of a digital acoustic signal and an impulse
response stored in storage means 35, according to the digital address signal of address signal
conversion means 34 By specifying the address of the storage unit 35 and reading out the
impulse response digitally recorded in the storage unit 35, the digital acoustic signal is corrected
in real time based on the head movement of the listener 23 based on this impulse response. It is
[0015]
Further, as shown in FIG. 1, the audio reproducing apparatus according to the present invention
comprises an analog signal source 2 for supplying analog audio signals of a plurality of channels,
and an analog / digital converter for converting analog audio signals from the analog signal
source 2 into digital audio signals. Sound-reproducing means 24 arranged in the vicinity of the
ears of the listener 23 and the listener 23 and converting the digital sound signal from the
analog-digital converting means 3 into an analog sound signal by the digital-analog converting
means and reproducing it; Angle detection means 28 for detecting the head movement of the
listener 23 with respect to the reference direction every predetermined angle; address signal
conversion means 34 for converting the angle detected by the angle detection means 28 into a
digital address signal; From the virtual sound source position relative to the reference direction
of the head of the listener 23 for each distinguishable angle, both Integrating means 5, 6, 7, 8 for
performing convolutional integration of the digital acoustic signal from the analog / digital
converting means 3 and the impulse response stored in the storing means 35. 9, 10, 11, and 12,
the address of the storage means 35 is designated by the digital address signal of the address
signal conversion means 34, and the impulse response digitally recorded on the storage means
35 is read out, based on this impulse response. Digital acoustic signals based on the head
movement of the listener 23 in real time.
[0016]
Further, as shown in FIG. 1, the audio reproducing apparatus of the present invention comprises
an analog signal source 2 for supplying analog audio signals of a plurality of channels, a digital
signal source 1 for supplying digital audio signals of a plurality of channels, and an analog signal
source 2 An analog / digital conversion means 3 for converting an analog audio signal from the
digital audio signal into a digital audio signal, a switching means 4 for switching between the
digital audio signal from the digital signal source 1 and the digital audio signal from the analog /
digital converter 3; Sound reproducing means 24 disposed in the vicinity of the both ears of the
person 23, converting the digital sound signal from the switching means 4 into an analog sound
signal by digital / analog converting means and reproducing it, and the head of the listener 23
10-05-2019
5
with respect to the reference direction By means of angle detection means 28 for detecting
movement at predetermined angles, and angle detection means 28. Address signal converting
means 34 for converting the emitted angle into a digital address signal, and for every angle that
the listener 23 can identify, from the virtual sound source position to the reference direction of
the head of the listener 23 to both ears of the listener 23 Integral means 5, 6, 7, 8, 9, 10, 11 for
performing convolutional integration of the digital sound signal from the switching means 4 with
the digital acoustic signal from the switching means 4 and the impulse response stored in the
memory means 35. , 12, the address of the storage means 35 is designated by the digital address
signal of the address signal conversion means 34, and the digital acoustic signal is read based on
the impulse response by reading out the impulse response digitally recorded in the storage
means 35. Correct in real time based on the head movement of the listener 23 It is obtained by
the.
[0017]
Further, as shown in FIG. 1, the audio reproducing apparatus according to the present invention
comprises a digital signal source 1 for supplying digital audio signals of a plurality of channels,
and a digital signal from the digital signal source 1 disposed near the listener's 23 ears. An
acoustic reproduction means 24 for converting an acoustic signal into an analog acoustic signal
by digital / analog conversion means and reproducing the same; an analog angle detection means
38 for analog detection of head movement of the listener 23 with respect to a reference direction
at predetermined angles; Analog / digital conversion means 43 for converting analog signals
detected by the angle detection means 38 into digital angle signals, and address signal
conversion for converting angles as digital signals converted by the analog / digital conversion
means 43 into digital address signals Means 34 and an angle which can be identified by the
listener 23 Means 35 for digitally recording an impulse response from the virtual source position
to the reference direction of the head of the person 23 to the ears of the listener 23 and the
acoustic signal from the digital signal source 1 and the impulse response stored in the means 35
, Integrating means 5, 6, 7, 8, 9, 10, 11, 12 for performing convolutional integration with each
other, the address of the storage means 35 is designated by the digital address signal of the
address signal conversion means 34, By reading out the digitally recorded impulse response, the
digital acoustic signal is corrected in real time based on the head movement of the listener 23
based on the impulse response.
[0018]
According to the present invention, the convolutional integration of the digital acoustic signal
from the digital signal source 1 and the impulse response at each angle that can be identified by
the listener 23 digitally stored in the storage means 35 is performed 5, 6, 7; 8, 9, 10, 11, 12,
10-05-2019
6
high-speed processing of digital sound signals can be corrected in real time based on the head
movement of the listener 23, and the capacity of the storage means 35 can be reduced.
Further, according to the present invention, the analog acoustic signal from the analog signal
source 2 is converted into a digital acoustic signal, and the digital acoustic signal and the impulse
response for each angle that can be identified by the listener 23 digitally stored in the storage
means 35 5, 6, 7, 8, 9, 10, 11, 12, etc., which perform convolutional integration, can be corrected
in real time based on the head movement of the listener 23 by processing digital audio signals at
high speed, and storing The capacity of the means 35 can be reduced and the capacity of the
storage means 35 can be reduced.
[0019]
Further, according to the present invention, the digital acoustic signal from the digital signal
source 1 and the analog acoustic signal from the analog signal source 2 are converted into digital
acoustic signals to be digital acoustic signals, and these two digital acoustic signals are switched.
5, 6, 7, 8, 9, 10, 11 and 12 which perform the convolution integration of the digital acoustic
signal and the impulse response for each angle that can be identified by the listener 23 digitally
stored in the storage means 35 by switching. The digital acoustic signal can be corrected at high
speed based on the head movement of the listener 23 in real time, and the capacity of the
storage means 35 can be reduced.
[0020]
Further, according to the present invention, the head movement of the listener 23 with respect to
the reference direction is detected 38 at a predetermined angle, converted into an analog signal
43, and converted 43 into a digital angle signal. 5, 6, 7, 8, 9, 10, which reads out impulse
responses from the storage means 35 and performs convolutional integration of digital audio
signals and impulse responses for each angle that can be identified by the listener 23 digitally
stored in the storage means 35 Since the digital acoustic signal is processed at high speed, the
digital acoustic signal can be corrected in real time based on the head movement of the listener
23, and the capacity of the storage means 35 can be reduced.
[0021]
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT An embodiment of an audio
reproducing apparatus according to the present invention will be described in detail with
reference to FIGS.
10-05-2019
7
The audio reproducing apparatus according to the embodiment of the present invention, when
reproducing an audio signal with headphones, localization equivalent to reproduction of sound
from the speaker that should be placed in a predetermined positional relationship when
reproducing with the speakers. The feeling, the feeling of the sound field, etc. can be obtained
even by reproducing with headphones.
[0022]
That is, the audio reproducing apparatus according to the embodiment of the present invention is
used in a system for reproducing multi-channel audio signals collected by stereo or the like with
headphones.
In particular, a predetermined positional relationship (for example, front right, front left, center,
etc. of the listener).
The digital audio signal recorded or transmitted in each channel for the purpose of localizing
each sound image) is used when reproducing with a headphone or the like.
[0023]
Here, head movement relative to the reference direction is first detected at a constant angle or at
a predetermined angle, and converted into a digital address signal representing a magnitude
including the direction.
By means of this address signal, the digitally recorded impulse response from the virtual sound
source position to the reference direction of the head prerecorded in the memory to both ears is
read out.
Then, convolution integration of the digitized acoustic signal of each channel and this impulse
response is performed to correct and change in real time.
10-05-2019
8
Thereby, it is possible to realize a reproduction effect as if the reproduction sound can be heard
from the speaker placed at the virtual sound source position.
[0024]
In FIG. 1, reference numeral 1 denotes a multi-channel digital stereo signal source such as a
digital audio disc (for example, a compact disc) and digital satellite broadcasting.
The code ¦ symbol 2 shows analog stereo signal sources, such as an analog record and an analog
broadcast.
Reference numeral 3 is an A / D converter for converting these analog signals into digital signals.
This A / D converter 3 is provided for the number of channels in the case of multiple channels.
Reference numeral 4 denotes a switch, and a digital input signal and an analog input signal are
treated as digital signals represented by a constant sampling frequency and the number of
quantization bits. Here, only switching of two channels is shown, but in the case of multiple
channels, the same number of channels is provided.
[0025]
The left digital signal L out of these digital signal sequences is supplied to the convolution
integrator 5. Here, a constant sampling frequency from the virtual sound source position to both
ears with respect to the reference direction of the head in the direction in which the head of the
listener 23 currently faces is called to the memory 6 attached to the convolutional integrator 5
And convolutional integration is performed with a set of digitally recorded impulse responses
represented by and the number of quantization bits. Also, the convolution integrator 7 and the
memory 8 supply the crosstalk component of the right digital signal R.
[0026]
10-05-2019
9
Similar to the above, the right digital signal R is supplied to the convolution integrator 11. Here, a
constant sampling frequency from the virtual sound source position to both ears with respect to
the reference direction of the head in the direction in which the head of the listener 23 currently
faces is called to the memory 12 attached to the convolutional integrator 11 And convolutional
integration is performed with a set of digitally recorded impulse responses represented by and
the number of quantization bits. Also, the convolution integrator 9 and the memory 10 supply
the crosstalk component of the left digital signal L.
[0027]
The result of convolution integration of the impulse response and real time in the convolutional
integrator 5 and the memory 6, the convolutional integrator 9 and the memory 10 is supplied to
the adder 15 and added. Also, the result of convolution integration of impulse response and real
time in the convolutional integrator 7 and the memory 8, the convolutional integrator 11 and the
memory 12 is supplied to the adder 16 and added. At this time, reverberation signals from the
reverberation circuits 13 and 14 are added to the adders 15 and 16.
[0028]
The result of the convolution integration and addition by the adders 15 and 16 is corrected by
the correction circuits 17 and 18 in order to remove characteristics specific to the headphones
used to measure the impulse response, and then the D / A converter 19 The signal is converted
into an analog signal by 20, power-amplified by the power amplifiers 21, 22 and added to the
headphone 24. Thereby, the listener 23 can listen to the sound by the right speaker and the left
speaker 26 of the headphone 24.
[0029]
The correction circuits 17 and 18 are provided in any part of the path from the acoustic signal
being added by the multi-channel digital stereo signal source 1 and the multi-channel analog
stereo signal source 2 to the various signal processing and being added to the headphones 24. It
may be either analog signal processing or digital signal processing.
[0030]
In convolution integration with a set of digitally recorded impulse responses from the virtual
10-05-2019
10
source position to both ears of the listener 23 with respect to the reference direction in the
convolutional integrators 5, 7, 9, 11 and the memories 6, 8, 10, 12 Alternatively, the impulse
response of the correction coefficient of the headphone 24 may be pre-folded into an impulse
response to perform convolution processing including correction simultaneously in real time.
[0031]
Here, the digital angle detector 28 detects the movement of the head of the listener 23, and the
detailed configuration of the digital angle detector 28 is shown in FIG.
FIG. 2 shows the case where the horizontal component of geomagnetism is used as the digital
angle detector 28.
FIG. 2 shows an example in which the angle detection signal is extracted as a digital signal.
[0032]
First, as an example in the case of extracting the head movement of the listener 23 with respect
to the reference direction as discrete information for each fixed unit angle or for each
predetermined angle, the rotary encoder 30 sets the input axis to be vertical at the head center
position. And the magnetic needle 29 is provided on the input shaft. Therefore, from the rotary
encoder 30, an output indicating the movement of the head including the direction of the listener
23 is taken with reference to the north-south direction indicated by the magnetic needle 29.
Although this rotary encoder 30 is attached to the headband 27 of the headphone 24, it may be
provided on an attachment device independent of the headband 27.
[0033]
Then, the output of the encoder 30 of the digital angle detector 28 is supplied to the detection
circuits 31 and 32, and when the listener 23 turns the head clockwise and when the listener 23
turns it counterclockwise, the detection circuit 31 The direction signal Sd changing to 0 "or" 1 "is
taken out, and when the listener 23 changes its head direction, the detection circuit 32 generates
Pa, for example, 2 ° in number of pulses proportional to the changed angle. One pulse Pa is
output each time it changes.
10-05-2019
11
[0034]
Then, the signal Sd is supplied to the count direction input U / D of the up / down counter 33,
and the pulse Pa is supplied to the clock input (count input) CK of the up / down counter 33. It is
converted into a digital address signal representing the direction and size of the signal, and is
supplied to the memory 35 as an address signal through the address control circuit 34.
[0035]
Then, from the corresponding address of the table in the memory 35, a digitally recorded
impulse response from the virtual sound source position to the reference direction of the head of
the listener 23 previously recorded in the memory 35 to both ears of the listener 23 is read At
the same time, the convolutional integration of this impulse response with the digitized acoustic
signal of each channel, loaded in the memories 6, 8, 10, 12 attached to the convolutional
integrators 5, 7, 9, 11 is performed, Correction in the direction in which the head of the listener
23 is directed is performed in real time.
[0036]
Reference numeral 38, on the other hand, indicates an analog angle detector, the detailed
configuration of which is shown in FIG.
FIG. 3 shows an example in which the angle detection output is taken out as an analog signal.
A light receiver 41 composed of a light receiving element whose resistance value changes
according to the intensity of light such as a CDS or a photodiode is attached to the central
portion of the head of the listener 23.
A light emitting device 39 such as a light bulb or a light emitting diode is provided opposite to
the light receiving device 41, and the light emitting device 39 irradiates the light receiving device
41 with light of a certain intensity.
[0037]
10-05-2019
12
At this time, a movable shutter 40 is provided between the paths of projected light of the light
emitter 39 so that the degree of transmission of the projected light changes with the rotation
angle, and the movable shutter 40 rotates with the magnetic needle 29. There is. Therefore, when
a constant current is supplied to the light receiver 41, the voltage at both ends of the light
receiving element of the light receiver 41 takes an analog output indicating the movement of the
head including the direction of the listener 23 based on the north-south direction indicated by
the magnetic needle 29. Ru. The analog angle detector 38 is attached to the headband 27 of the
headphone 24 but may be provided on an attachment device independent of the headband 27.
[0038]
The analog output of the analog angle detector 38 is amplified by the amplifier 42 and then
applied to the A / D converter 43, and the digital output is supplied to the address control circuit
34 via the switch 44. The address control circuit 34 generates a digital address signal
representing the magnitude of head movement of the listener 23 with respect to the reference
direction including the direction at a constant angle or a predetermined angle, and is supplied to
the memory 35 as an address signal.
[0039]
Then, from the corresponding address of the table in the memory 35, a digitally recorded
impulse response from the virtual sound source position to the reference direction of the head of
the listener 23 previously recorded in the memory 35 to both ears of the listener 23 is read At
the same time, the convolutional integration of this impulse response with the digitized acoustic
signal of each channel, loaded in the memories 6, 8, 10, 12 attached to the convolutional
integrators 5, 7, 9, 11 is performed, Correction in the direction in which the head of the listener
23 is directed is performed in real time.
[0040]
FIG. 4 shows an example of data of a table in the memory 35.
That is, now, as shown in FIG. 5, when the left front and right front speakers 45L, 45R are
disposed in front of the listener 23, from the installation position of the left and right speakers
45L, 45R, As impulse response to reach both ears
10-05-2019
13
[0045] When considering the table of the memory 35, digitally recorded impulse responses
representing these are recorded. Hmn (t, θ): Impulse response from m speaker position to n
ears. θ: m Speaker position and head-formed angle Hmn (ω, θ): Transfer function from m
speaker position to n ears. ω: An angular frequency (2πf, f: frequency), and a speaker may be
used as a sound source for measuring an impulse in this case. Further, the sound collection
position of each ear of the listener 23 may be any position from the entrance of the ear canal to
the tympanic membrane position.
[0046] However, this position is required to be equal to a position for obtaining a correction
characteristic for canceling the inherent characteristic of the headphone used for measuring the
impulse response described later. When such an impulse response is considered, the digitally
recorded impulse response when the angle θ is changed, for example, by 2 ° at each unit angle
is written for each address 1 of the table of the memory 35. This angle is set to an angle that
allows the left and right ears to identify the rotated angle of the head when the listener rotates
the head. Further, for example, three sets of this table are provided for the memory 35, and for
each set, the shape of the head and pinna of the listener 23, the characteristics of the
headphones used for measuring the impulse response, etc. The values are made to be different.
Then, one of the three sets of tables is selected in accordance with the switching of the switch 36
of the address control circuit 34.
[0047] In FIG. 1, reference numeral 37 denotes a reset switch. When the reset switch is turned
on, the value of the up / down counter 33 is reset to "all 0", and at this time, an address of θ = 0
is selected in the table of the memory 35. .
[0048] The audio reproducing apparatus of this embodiment is configured as described above
and operates as follows. A digital audio signal from multi-channel digital stereo signal source 1 or
an audio signal of each channel obtained by converting an analog signal input to multi-channel
analog stereo signal source 2 into a digital signal by A / D converter 3 is switch 4 After being
selected, the digital data to both ears with spatial information as a sound field by the
convolutional integrators 5, 7, 9, 11 and the circuits of the memories 6, 8, 10, 12 and the adders
15, 16 The signal is converted to a signal, is converted to an analog signal by the D / A
converters 19 and 20 through the correction circuits 17 and 18 of headphones used for
measurement of impulse response, and power amplified by the power amplifiers 21 and 22.
Supplied.
[0049] Then, in this case, when the listener 23 moves the head, in the case of using the digital
10-05-2019
14
angle detector 28 as the digital angle detection means, the signals Sd and Pa according to the
direction are obtained. The count value is a value according to the direction of the head of the
listener 23. The count value is supplied to the memory 35 as an address signal through the
address control circuit 34. From the memory 35, among the data corresponding to the table of
FIG. 4, the digitally recorded impulse response from the virtual sound source position to the both
ears with respect to the reference direction of the head corresponding to the head direction of
the listener 23 is taken This data is provided to convolutional integrators 5, 7, 9, 11 and
memories 6, 8, 10, 12.
[0050] In the case of using the analog angle detector 38 for detecting head rotation, the sensor
output is amplified by the amplifier 42 and then converted by the A / D converter 43 into a
digital signal according to the head orientation of the listener 23 It is supplied as an address
signal to the memory 35 through the control circuit 34 and digitally recorded from the virtual
sound source position to both ears with respect to the reference direction of the head
corresponding to the head direction of the listener 23 in the same manner as the digital angle
detector 28. The impulse response is taken and this data is provided to the convolutional
integrators 5, 7, 9, 11 and the memories 6, 8, 10, 12.
[0051] Therefore, the audio signals L and R supplied to the headphone 24 are subjected to
convolutional integration with the digitally recorded impulse response from the virtual sound
source position to the both ears with respect to the reference direction of the head
corresponding to the head direction of the listener 23 As a result, a plurality of speakers can be
placed at the virtual sound source position, and a sound field feeling as reproduced by the
speakers can be obtained.
[0052] In particular, according to this example, when changing the characteristics of the audio
signal supplied to the headphones 24 in accordance with the head orientation of the listener 23,
this is performed using the table of the memory 35. It can be changed finely according to the
direction of the head, and it can be made an optimal characteristic.
[0053] Furthermore, since the impulse response digitally recorded in the table of the memory 35
is taken out and this data is supplied purely electronically to the convolutional integrators 5, 7, 9,
11 and the memories 6, 8, 10, 12, the listener There is no delay in the change in the
characteristics of the audio signal with respect to the orientation of the head 23 and no
unnaturalness.
10-05-2019
15
[0054] At this time, since the reverberation signals from the reverberation circuits 13 and 14 are
also supplied to the headphones 24, a sense of expansion in the listening room or a concert hall
is added, and an excellent sense of stereo sound field can be obtained.
[0055] On the other hand, a plurality of tables are prepared in the memory 35 so that the
listener 23 can arbitrarily select them by the switch 36. Therefore, the shape of the head and the
pinna of the listener 23, the characteristics of the headphones used Even if they are different,
optimum characteristics can be obtained. Furthermore, the amount of change in the digitally
recorded impulse response from the virtual source position to both ears with respect to the
reference direction of the head of the listener 23 with respect to the change of the angle θ is
made larger or smaller than the standard value by the table. As a result, the amount of change in
the position of the sound image with respect to the direction of the head of the listener 23
differs, so that the sense of distance from the listener 23 to the sound image can be changed.
[0056] Moreover, while adding the reverberation signal by the reverberation circuit 13 and 14,
this reverberation signal sounds like reflection sound and reverberation by the wall of a hole etc.,
so it seems as if listening to music in a famous concert hall You can get a sense of reality.
[0057] Further, the data of FIG. 4 can be obtained as follows. That is, the impulse sound source
having the required number of channels and the dummy head microphone are disposed at
predetermined positions in an appropriate room so as to be a preferable reproduction sound field
when reproduced by the headphones 24. A speaker may be used as a sound source for
measuring the impulse in this case.
[0058] With regard to the sound collection position of each ear of the dummy head, any position
between the entrance of the ear canal and the tympanic membrane position may be used, but a
position where the correction characteristic for canceling the characteristic peculiar to the
headphones used for measuring impulse response It is required to be equal.
[0059] Further, the measurement of the impulse response can be obtained by emitting impulse
sound from the speaker position of each channel and collecting the sound with a microphone
provided for each ear of the dummy head at every constant angle: Δθ. Therefore, at a certain
angle: θ 1, one set of impulse responses is obtained for each channel, so in the case of a 5channel signal source, 5 sets for each angle, that is, 10 types of impulse responses Will be
obtained.
10-05-2019
16
[0060] In addition, the method of finding the correction characteristic for canceling the
characteristic peculiar to the headphone used for the measurement of the impulse response is to
use the same headphone as the dummy head microphone which collected the impulse response
of the sound field. Wear it, and calculate the impulse response between the headphone input and
the microphone of each ear of the dummy head and the impulse response that is the inverse
characteristic of the result.
[0061] Alternatively, it may be obtained directly using adaptive processing such as LMS
algorithm. The correction of specific headphone-specific characteristics is an impulse response
that represents the calculated correction characteristics as processing in the time domain in an
arbitrary part from when the audio input signal is added to when the signal is added to the
headphone. And, after analog-to-digital conversion, they can be realized by passing through an
analog filter with an inverse characteristic.
[0062] Furthermore, in the above description, only the orientation of the head of the listener 23
in the horizontal plane is considered, but the orientation in the vertical plane and in the plane
orthogonal to these can be processed in the same manner.
[0063] Further, the table in the memory 35 can be one set, and the address control circuit 34
can change the designation of the address to the table to obtain control data as in the case where
there are a plurality of sets of tables.
[0064] Furthermore, the data of the table may be limited to the general head direction range of
the listener 23, and the angle θ is set, for example, every 0.5 ° around θ = 0 °, ¦ The interval
of the angle θ may be made different depending on the direction, such as setting every 3 ° for
θ4545 ° ¦. As described above, it may be every angle at which the listener can identify the
angle of head rotation. Furthermore, instead of the headphones 24, speakers arranged near the
listener 23 may be used.
[0065] According to the above example, the input audio signal can be adapted to either a digital
recorded or transmitted signal collected by multi-channel stereo or the like, or an analog
recorded or transmitted signal, and a listener The angle detection means for detecting the
movement of the head 23 can be applied to both of those output as digital signals and those
output as analog signals.
10-05-2019
17
[0066] Also, according to the above example, when changing the characteristics of the audio
signal supplied to the headphones 24 in synchronization with the movement of the head of the
listener 23, it is not continuous to the movement of the head of the listener 23 but the human
hearing. Since it is performed by reading from the table of the memory 35 every appropriate and
appropriate unit angle that can be identified by the human being or according to the
characteristic according to the characteristics, the head 23 of the listener 23 is necessary and
sufficient If the operation is performed only on the change contents, the same effect as
continuous change can be obtained. Therefore, the capacity of the memory 35 can be saved, and
the processing speed of the operation is not required to be unnecessarily high speed.
[0067] Further, according to the above example, since the binaural characteristic from the fixed
sound source in a fixed direction is always obtained regardless of the rotation of the head of the
listener 23, it is possible to obtain an extremely natural sense of out-of-head localization.
[0068] Also, according to the above example, according to the table of the memory 35, the
characteristic represented by the digitally recorded impulse response can be obtained by the
purely electronic formula in the convolutional integrators 5, 7, 9, 11 and the memories 6, 8, 10,
12. Since it performs convolutional integration at the same time, there is little characteristic
degradation, and no delay is caused in the change of the characteristic of the audio signal with
respect to the head movement of the listener 23. This causes unnaturalness as in the
conventional system. There is nothing to do.
[0069] Further, according to the above example, since a plurality of tables are prepared in the
memory 35 and the listener 23 can arbitrarily select them by the switch 36, the shape of the
head and the pinna of the listener 23 , Even if the characteristics of the headphones 24 are
different, optimum characteristics can be obtained.
[0070] Further, according to the above example, the change of the position of the sound image
with respect to the head direction of the listener 23 is made by setting the change amount of the
impulse response to the change of the angle θ larger or smaller than the standard value by the
table. Since the amounts are different, the sense of distance from the listener 23 to the sound
image can be changed.
[0071] Also, according to the above example, since appropriate reverberation signals are added
as needed by the reverberation circuits 13 and 14, it is possible to obtain a sense of presence as
10-05-2019
18
if listening to music in a famous concert hall.
[0072] According to the present invention, the digital acoustic signal from the digital signal
source and the impulse response for each angle which can be identified by the listener digitally
stored in the storage means are subjected to convolution integration, so that the digital acoustic
signal can be processed at high speed. The processing can be corrected in real time based on the
listener's head movement, and the storage capacity of the storage means can be reduced.
[0073] Further, according to the present invention, an analog acoustic signal from an analog
signal source is converted into a digital acoustic signal, and the digital acoustic signal and a
convolution integral of an impulse response at each angle that can be identified by a listener
digitally stored in storage means are identified. Therefore, by processing digital audio signals at
high speed, correction can be made in real time based on the head movement of the listener, and
the storage capacity of the storage means can be reduced.
[0074] Further, according to the present invention, a digital audio signal from a digital signal
source and an analog audio signal from an analog signal source are converted into a digital audio
signal to obtain a digital audio signal, and the two digital audio signals are switched to digital
Since the convolutional integration of the acoustic signal and the impulse response for each
angle that can be identified by the listener digitally stored in the storage means is performed, the
digital acoustic signal is processed at high speed to correct in real time based on the head
movement of the listener. Storage capacity of the storage means can be reduced.
[0075] Further, according to the present invention, the head movement of the listener with
respect to the reference direction is detected at predetermined angles and converted into an
analog signal, which is converted into a digital angle signal, and from the storage means The
impulse response is read out and convolution integration of the digital acoustic signal and the
impulse response for each angle that can be identified by the listener digitally stored in the
storage means is performed. The correction can be made in real time based on the above, and the
storage capacity of the storage means can be reduced.
10-05-2019
19

Download Report