close

Вход

Забыли?

вход по аккаунту

?

JP2018518880

код для вставкиСкачать
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 JP2018518880
Abstract: An acoustic diffusion manifold transducer system comprising a plurality of (N or N2)
acoustic channels (where N is an odd prime number) arranged in an N × 1 or N × N matrix,
each acoustic channel Are driven by the loudspeaker driver, and each channel length is related
by Ti. j = [(i 2 + j 2) rem N] <*> determined by unit delay. Where T is the delay between channels
having sequence values in the number sequence and N is a prime number. The channels are
arranged to terminate the outlet so that sound waves from the speaker driver arrive in a fixed
order. The outlet of each channel has the same area. The channel is a path for the sound waves
generated by the loudspeaker driver, preferably a closed tube with any suitable cross section.
Preferably, the cross-sectional area of each path is identical but the path length is determined by
an algorithm. Preferably, the number sequence used in the acoustic diffusion manifold is selected
from a square residue sequence, a Barker code, an autocorrelation sequence, or a complementary
sequence. [Selected figure] Figure 1
Sound diffusion manifold
[0001]
The present invention relates to an acoustic arrangement, in particular to an acoustic
arrangement providing means for generating diffuse waves in a fluid space. More particularly,
the present invention relates to a loudspeaker configuration adapted to generate diffuse waves.
[0002]
02-05-2019
1
WO 2012015650 discloses reflectors and other configurations that generate diffusive waves in
fluid space to clarify energy and enhance specific information in the space carrying audio signals.
In part, a manifold is briefly disclosed.
[0003]
Some loudspeaker drivers accelerate the movement of the apparent sound center significantly at
very high frequencies. The acoustic center starts moving towards the driver's voice coil faster, for
example at about 10 kHz.
[0004]
The design of some of the embodiments of the acoustic reflector disclosed in WO2012015650
was sensitive to the geometric movement of the acoustic center and had to be adapted to such
movement.
[0005]
The object of the present invention is to provide an improvement of the invention disclosed in
WO2012015650.
[0006]
The present invention comprises a plurality of (N or N2) acoustic channels (where N is an odd
prime number) faces arranged in an N × 1 or N × N matrix, each acoustic channel being driven
by a loudspeaker driver And each channel length is related Ti. Provided is an acoustic diffusion
manifold transducer system, determined by j = [(i 2 + j 2) rem N] <*> unity delay, where T is the
delay between channels having series values in the sequence and N is a prime number.
[0007]
The channels are arranged to be terminated at the outlet device so that sound waves from the
speaker driver arrive in a fixed order.
The outlet of each channel has the same area.
02-05-2019
2
The channel is a path for the sound waves generated by the loudspeaker driver, preferably a
closed tube with any suitable cross-sectional area.
Preferably, the cross-sectional area of each path is identical but the length of the paths is
determined by an algorithm that achieves diffusion.
[0008]
Preferably, the number sequence used in the acoustic diffusion manifold is selected from a
square residue sequence, a Barker code, an autocorrelation sequence, or a complementary
sequence.
[0009]
Other suitable sequences are those used in signal processing such as Barker codes, zero
autocorrelation sequences, or complementary sequences.
[0010]
The Barker code is a sequence of N values consisting of +1 and -1, and
[0011]
Autocorrelation is the cross correlation between a signal and the signal itself.
Simply stated, autocorrelation is the similarity between measurements as a function of the time
interval between measurements.
Autocorrelation is a mathematical tool to find repetitive patterns such as the presence of periodic
signals embedded in noise or to identify missing fundamental frequencies in the signal indicated
by its harmonic frequency.
Autocorrelation is often used in signal processing to analyze a series of values, such as a function
02-05-2019
3
or time domain signal.
[0012]
Complementary sequences (CS) are derived from applied mathematics and are pairs of sequences
having the useful property that the sum of out-of-phase aperiodic autocorrelation coefficients is
zero. A complementary series consisting of two values is described by Marcel J. et al. E. It was
first announced by Golay in 1949. In 1961-1962, Golay presented several methods for
constructing 2N-long sequences, giving examples of complementary sequences of 10 and 26 in
length. In 1974, R.S. J. Turyn presents a method for constructing a sequence of length mn from a
sequence of length m and a sequence of length n, this method is arbitrary in the form 2 <N> 10
<K> 26 <M> It enabled construction of a series of lengths.
[0013]
FIG. 5 is an isometric view of an acoustic manifold. FIG. 10 is an isometric view of a loudspeaker
driver and an acoustic diffuser manifold. FIG. 2 is a plan view and an elevation view of the
acoustic diffusion manifold shown in FIG. 1; FIG. 5 is a bottom view of the hard on collider
loading area. FIG. 5 is a plan view that quickly identifies data points of concentric splitters
located in a hard on collider. FIG. 5 is a representative view showing both the hard on collider
portion of the manifold and the radiating outlet portion. FIG. 7 is a detailed isometric view of the
“twister” component. FIG. 6 shows a cross-sectional slice of a portion of a twister element. FIG.
5 shows an uncompressed channel design. FIG. 5 shows a compressed channel design. FIG. 10 is
an isometric view of a miniature acoustic diffusion manifold. FIG. 5 is an isometric view showing
a skeleton of a miniature acoustic diffusion manifold. FIG. 6A is a top view of the section element
path of the miniature acoustic manifold. FIG. 5 shows the lower layer 401 of a miniature acoustic
diffusion manifold. FIG. 5 illustrates the size of the outlet of the Cobra manifold. FIG. 7 illustrates
the hard on collider branch area of the Cobra manifold. FIG. 5 is a detailed view of the captured
impulse response of the loudspeaker of the invention. FIG. 5 illustrates a fast Fourier transform
FFT of the manifold of the present invention. FIG. 7 is an isometric view of a manifold
loudspeaker with wavelet transient ring radiation. FIG. 7 shows a listener associated with a single
manifold loudspeaker emitting a wavelet ring. FIG. 6 shows a stereo space of manifold
loudspeakers emitting individually different wavelet ring patterns. FIG. 1 shows a complete
surround sound system using three manifold speakers. FIG. 5 illustrates an expanded virtual
space environment in which five manifold loudspeakers are used. FIG. 5 is an isometric view of a
manifold speaker driver configuration. FIG. 1 is an isometric view of a car dashboard including
two manifold speaker configurations. FIG. 5 is an isometric view of a manifold speaker driver.
02-05-2019
4
FIG. 1 is a front isometric view of a flat screen TV. Figure 14 is an isometric view of the back of
the flat screen TV. FIG. 5 graphically illustrates tones and their fast Fourier transforms. FIG. 5
graphically illustrates tones and their fast Fourier transforms. FIG. 5 graphically illustrates tones
and their fast Fourier transforms. FIG. 5 is a schematic diagram of a sharp phased signal injection
system based on bass energy in the stop band of a loudspeaker.
[0014]
FIG. 1 is an isometric view of an acoustic diffusive manifold 101 with seven zones 102, with two
zone pairs interchanged with one another through the twister section 103, resulting in all seven
zones 0-6 (7 elements). It reaches outlet 104 in a continuous series.
[0015]
The length of each interval 102 is determined by the solution of the squared residue series, and a
fixed offset distance of Added to the length.
[0016]
The solution for QRD is determined so that the relative length variation between the intervals
falls at the outlet 104 in the series 2, 4, 1, 0, 1, 4, 2.
The relative positions within the natural seven sections of the hard-on collider are 4, 2, 1, 0, 1, 2,
4.
Thus, it is essential that the outer elements representing the "2" and "4" elements alternate along
the path from the hard on collider to the outlet.
[0017]
FIG. 2 is an isometric view showing the coupled position of the two components, the loudspeaker
driver and the acoustic diffusion manifold. The manifold 201 is connected to the driver 202 at a
splitter intake 203 called a hard on collider. The role of the "hard on collider" is to equally orient
the acoustic waves generated by the piston movement of the loudspeaker driver to seven (or N),
but the individual acoustic waves travel along the sectional channel. This should be achieved
02-05-2019
5
without causing distortion or reflection of the acoustic energy. Therefore, in defining this part,
keeping the cross-sectional area and the general acoustic wave induction design method should
be considered.
[0018]
One of the mathematical sequences that can generate a response of the spreading wave with
autocorrelation equal to zero is known as a square residue sequence (QRS). QRS can be any odd
prime number N (e.g., 1, 3, 5, 7, 11, 13, 17, 19, 23, 29, ... N is the number of sections in the
manifold, with a total element length equal to. The solution of each element is determined by the
relation Sn = n <2> remN (ie, the minimum non-negative remainder that is the result of
subtracting n <2> from multiples of N).
[0019]
Table 1 shows the solution of QRS derived for a sequence with 7 elements (ie N = 7).
[0020]
It is the nature of the QRS that the spreading wave function can be realized using any one time
period (N adjacent elements) of the sequence.
Thus, the sequence starts with an arbitrary number n or its variance, ie as long as Nw in the
periodic width (where w is the width of the recess) is the solution of one complete period of the
sequence Can. Table 2 below starts with n = 4 and includes n = 10, i.e. N = 7 elements.
[0021]
Table 3 below starts with n = 2 and contains n = 6, i.e. N = 5 elements. Solutions 4, 1, 0, 1, 4
appear nested within solutions 2, 4, 1, 0, 1, 4, 2 of Table 2. It is the nature of QRS that small
prime solutions appear nested within large prime solutions.
[0022]
02-05-2019
6
If an arbitrary set of N solutions Sn is not suitable for the application, then a constant can be
added to each solution Sn and applied to the formula Sn = (Sn + a) rem N, where a is a constant.
[0023]
Thus, if N = 7 eigen solutions are 0, 1, 4, 2, 2, 4, 1, then for example a = 3 is added to each Sn
and the solution is 3, 4, 0, 5, 5, It can be converted to 0,4.
[0024]
FIG. 3 is a plan view and an elevation view of the acoustic diffusion manifold shown in FIG.
[0025]
The cross-sectional view of FIG. 3 along AA shows element 306 in detail, in which acoustic
energy is converted from the piston motion of the moving coil loudspeaker to horizontal motion
along the length of the section element.
[0026]
FIG. 4 is a bottom view of the concentric splitter inlet area 401 of the hard on collider of the
manifold shown in FIGS.
[0027]
This figure shows how a circular area divided into 7 (N) equal sections, which are hard on
colliders, divides the acoustic energy from the moving coil loudspeaker into 7 (N) equal area
parts In detail.
[0028]
Next, each equal part generated by the hard on collider enters the path for the segment elements
of different path lengths for individual guidance via the twister 303 and is acoustically sent 306
to the outlet 307.
[0029]
Some loudspeaker drivers accelerate the movement of the apparent sound center significantly at
very high frequencies.
02-05-2019
7
The acoustic center begins to move faster, for example at 10 kHz or more, towards the driver's
voice coil.
The manifold design incorporates a concentric splitter configuration in the hard on collider area
401 (FIG. 4).
This configuration is concentric with the loudspeaker driver and splits the acoustic drive wave
into N identical parts about the concentric axis, thus eliminating errors due to the movement of
the acoustic center.
Thus, changes in the acoustic center position exist symmetrically in all N intervals as long as they
are concentric with the driver on the path.
[0030]
The past design of the embodiment of the acoustic reflector by the inventor was susceptible to
the geometric movement of the acoustic center and had to be adapted to such movement.
[0031]
The manifold 301 of FIG. 3 has a plurality of channels whose length difference is a solution of
QRS × unit depth.
That is, the length of channel 0 (302) is 0 <*> unit length + constant l, and the length of channel
1 (303) adjacent to recess 0 (302) is 1 <*> unit depth + constant l And the length of the channel
2 (304) adjacent to the recess 1 (303) is 2 <*> unit depth + constant l.
Because the constant "l" is within the length of each channel, it does not form part of the
difference in length between the channels.
Components of acoustic energy emitted from source 302 (FIG. 2) are mixed in far-field space to
02-05-2019
8
exhibit a diffuse wave encoded sound field when emitted from outlet 307 having channels 302,
303, 304, 305 Is desirable.
The "perfect" solution of the QRS provides equal acoustic energy from the outlet 307 in all
angular directions, nominally within ± PI / 2 angular directions from the radial direction, and
actually at more angles.
[0032]
In the preferred practical design of an acoustic diffusion manifold suitable for a full range of
applications, the channel outlet width is selected to be 8.15 mm. Thus, the entire reflector is
57.05 mm.
[0033]
Typical QRD solutions are shown in Table 4 when the design frequency is chosen to be 2600 Hz.
Table 4
[0034]
The data to measure the interval channel length may be any suitable point on the hard on
collider region as long as the acoustic timing (phase) and amplitude are the same.
[0035]
FIG. 5 shows data points defined around the concentric splitter at the outer periphery of each
section.
Since this point is symmetrical in the section of the concentric splitter, the acoustic energy
present at each point can be considered to be temporally identical and can be considered as zero
data to each associated channel start point.
02-05-2019
9
[0036]
FIG. 6 is a representative view of both the concentric splitter and the radiating outlet portion of
the hard on collider portion of the manifold. Since the outlet is disposed on one side of the hard
on collider, the section portion is naturally dispersed in accordance with the series position of the
QRD element on the outlet.
[0037]
The closest section part is given the role of a "0" element, and the path length between the
section part and the outlet is set to the minimum. Generally, setting this distance to 0 mm is not
practical. Thus, the resulting distance is taken as a constant "l" which is added to the length of all
other element paths to add the set distance to all element paths. For example, using Table 4 as a
reference, the constant "l" is set to 50 mm. In practice, this length "l" can be in meters. Such long
constant lengths allow the driver to be located somewhat remote from the outlet. In this way, the
driver can be placed at the base of the flat screen TV and the radiation outlet can be placed at the
edge of the screen. Similarly, in a car, the driver can be embedded in the center of the dashboard
and the outlet can be on the surface of the dashboard.
[0038]
The hard on collider elements adjacent to the "0" section are assigned a "1" element path. Using
Table 4 as a reference, the length of this path is 69 mm, consisting of the constant "l" 50 mm and
the solution 19 mm of the "1" element.
[0039]
The path taken by the “1” element is configured such that the total travel to the outlet is a 69
mm translation in path length. Usually, the centerline of the interval channel path is considered
as the reference for measuring the length. An error caused by any acoustic phenomenon can be
corrected by compensating the error by actually increasing or decreasing the element length by
fine adjustment of the element path.
02-05-2019
10
[0040]
A section adjacent to the “1” element is assigned a “2” element path length. Using Table 4
as a reference, the "2" element path length is 88.2 mm, consisting of the constant "l" 50 mm and
the "2" element path length 38.2 mm.
[0041]
The path taken by the "2" element is configured such that the total travel to the outlet is a
translation of 88.2 mm in path length.
[0042]
The section adjacent to the “2” element is assigned the “4” element path length.
These two sections are adjacent to each other to complete seven sections of the hard on collider.
Using Table 4 as a reference, the "4" element path length is 126.2 mm, consisting of the constant
"l" 50 mm and the "4" element path length 76.2 mm.
[0043]
The path taken by the "4" element is configured such that the total travel to the outlet is a 126.2
mm translation in path length. However, the "2" and "4" elements must cross each other and
eventually be in the correct series in the outlet manifold.
[0044]
FIG. 7 is a detailed isometric view showing the “twister” components of segment elements
“2” and “4”, where the outer element 602 of the path length determined by “4” is a
constant cross-sectional area converter 601 Swap the position with the inner element 603 of the
path length determined by “2” via In this figure, the outer element 602 of the section element
formation is manipulated by the twister to be finally converted to the inner position before
reaching the outlet.
02-05-2019
11
[0045]
FIG. 8 shows seven sections passing through the twister section including the start point and the
end point. At the start, the separation fins are vertical and the area of the "4" element (A4) is
equal to the area of the "2" element (A2). In the next section, the center separation fin is starting
to rotate around the center point. The separation fins are longer and the width slightly shorter
than at the beginning. In this way, the correct cross-sectional area of A4 and A2 can be
maintained.
[0046]
In the next section, the separating fins cross over the vertical limit to the side walls. As a result of
the shortening of the length and the broadening of the width, the cross-sectional areas A4 and
A2 are maintained. This process continues in the remaining sections. This constant crosssectional area is maintained for channel and channel 2 as acoustic energy traverses the twister
section.
[0047]
The two main design variables, unit depth and element width, determine the useful frequency
bandwidth for which the acoustic diffusion manifold is useful. The minimum useful frequency is
controlled by the amount of path provided by the various recess depths. The maximum useful
frequency is controlled by the width of the recess. The acoustic energy does not travel the direct
path along the channel length if the associated wavelength is higher than the frequency equal to
2 × channel width. Because acoustic energy travels diagonally along the channel length, the
effective length is longer than the physical length. Because of this, the diffusion process exceeds
tolerances.
[0048]
In order to control the low frequency design frequency of the mechanical diffusion wave
generator, the unit length is set to be equal to 1 / N times the design wavelength. For example,
02-05-2019
12
for a unit length of 19 millimeters and N = 7, the design wavelength is given by: X=N×19ミ
リメートル=133ミリメートル
[0049]
From the above, the design frequency is calculated by: F = c / λD = 343 / (133 × 10 <-3>) = 2.6
kHz
[0050]
Below the design frequency, the depressions become dimensionally insufficient relative to the
phase of the source frequency, and the acoustic configuration acts as a normal radiator or planar
reflector. The highest frequency at which the reflector is effective, ie the cut-off frequency, is
determined by the relationship to the individual recess width w or the design frequency. Using
the previous example, if the recess width is 9.5 mm, then the cutoff frequency is given by: λ = w
× 2 = 19 mm
[0051]
Thus, the frequency is given by: F = c / λw = 343 / (19 × 10 <-3>) = 18.05 kHz
[0052]
Another factor that limits the effectiveness of high frequencies is that the sequence does not
function at (N-1) times the design frequency. That is, using the numbers of the previous example
also here: λhigh = λD / (N−1) λD = 133 mm Therefore, λhigh = 133 mm / 6 = 22.2 mm
Therefore, fhigh = 343 / λD = 343 / 22.2 mm = 15.5kHz
[0053]
In this example, the cutoff frequency determined by the design frequency is smaller than the
02-05-2019
13
lower of the two limit frequencies, which is the actual high frequency cutoff point. Thus, the
lower of the two frequencies is the cutoff frequency, ie 15.5 kHz.
[0054]
In order to protect against error interference due to the autocorrelation nature of the diffuse
wave function, a high degree of attention and accurate compensation must be incorporated into
the design. In zero autocorrelation, the output by itself does not convey meaningful information
that can be interpreted by a perceptive receptive organ such as a receptive organ of a human
listening system. The resulting diffuse wave function is "silence". However, because the
tolerances are very small, the percentage error from ideal should be less than 3% of the
amplitude or phase. The greater the error, the more audible the diffuse wave function. What we
want to hear in the listening space environment is not the diffuse wave function but the strength
of the drive source signal. Because QRS produces a wide range of frequencies, it is important to
maintain an error criterion nominally less than 3% at the top of the useful spectrum of the
design. As the frequency spectrum goes down, component waves increase, and if the spatial
origin of the source is stationary over the spectral region, the error due to the propagation of the
path is relatively small.
[0055]
In a preferred embodiment, the cross-sectional area of the hard on collider is the same as the
total area of the outlet. Efforts are made to make the cross-sectional area of each acoustic duct
constant from source to outlet.
[0056]
FIG. 9 shows an uncompressed cross-sectional structure where the area of the concentric splitter
section is identical to the cross-sectional area of the channel path.
[0057]
If the EG-concentric splitter diameter is 50 mm, the area of the concentric splitter is given by:
02-05-2019
14
AreaCS = PI × 25 mm <2> = 1963 mm <2> The area of one section is AreaSector = 1963 mm
<2> / 7 = 280 mm <2>
[0058]
If the channel width is 9.5 mm, the channel height is given by:
Height=AreaSector/Width =280/9.5 =29.5mm
[0059]
In another embodiment, the portion of the acoustic diffusion manifold that forms the hard on
collider (305 or FIG. 3) is an acoustic duct by compressing the cross sectional area of the section
outlet of the hard on collider bifurcation to the extent of the initial bifurcation area It is used to
amplify the volume velocity of the acoustic wave within. This raises the sound pressure level in
the duct. Area compression techniques should be used to take care that the acoustic waves do
not cause unwanted distortion.
[0060]
FIG. 10 shows a compressed cross-sectional structure where the area of the concentric splitter
section is larger than the cross-sectional area of the channel path.
[0061]
If the EG-concentric splitter diameter is 50 mm, the area of the concentric splitter is given by:
AreaCS = PI × 25 mm <2> = 1963 mm <2> The area of one section is AreaSector = 1963 mm
<2> / 7 = 280 mm <2>
[0062]
If the channel width is 9.5 mm, the channel height is given by:
02-05-2019
15
Height=AreaSector/(Width×Scale) =280/(9.5×2)
=14.8mm
[0063]
The outlet height was halved by introducing a scale factor of 2.
[0064]
As a result, the volumetric velocity of the acoustic energy in the channel can be predicted to be
twice that of the previous uncompressed configuration.
[0065]
The advantage of such an approach is that the size of the outlet manifold can be reduced so that
the resulting design can be compact.
[0066]
Figures 11-16 show a compact manifold suitable for use in a smart phone-type mobile phone.
[0067]
FIG. 11 is an isometric view of a compact acoustic diffusive manifold 1101 designed to
accommodate the Cobra smartphone loudspeaker 1102 through the recess 1103, the recess
1103 being divided into seven equal parts by a hard on collider, It is directed towards outlet
array 1104 by a variable length path determined by QRD.
[0068]
The loudspeaker driver is considered to behave as a complete piston over the frequency range of
application.
If this is not the case, a concentric splitter hard on collider branch configuration can be used.
[0069]
02-05-2019
16
If it is not preferable to couple the loudspeaker driver directly to the hard on collider area, small
air gaps can be used to elastically couple these elements.
Elastic space effects that absorb low frequencies are configured to occur below the effective
radiating portion of the smartphone's loudspeaker driver.
For Cobra loudspeakers, the effective radiation area is typically 500 Hz or higher.
Thus, the elastic gap should be an acoustic shortcut at 500 Hz or higher.
[0070]
FIG. 8 is an isometric view showing the skeleton of a compact acoustic diffuser manifold 701
without a case.
The various paths for the 7 (N) elements up to the outlet 704 are displayed. The aim is to make
the design as compact as possible without losing the desired sound effects provided by winding
the audio signal in the QRD.
[0071]
In order to control the low frequency design frequency of the compact acoustic diffusion
manifold, the unit length is set to be equal to 1 / N times the design wavelength. For example, if
the unit length is 15.5 mm and N = 7, then the design wavelength is given by:
X=N×15.5ミリメートル=108.6ミリメートル
[0072]
From the above, the design frequency is calculated by: F = c / λD = 343 / (108.6 × 10 <-3>) =
3.16 kHz
02-05-2019
17
[0073]
Below the design frequency, the recess is dimensionally insufficient relative to the phase of the
source frequency, and the acoustic configuration acts as a normal radiator or flat driver. The
highest frequency at which the reflector is effective, ie the cut-off frequency, is determined by the
relationship to the individual recess width w or the design frequency. Using the previous
example, if the recess width is 3.0 mm, then the cutoff frequency is given by: λ = w × 2 = 6.0
mm Thus, the frequency is given by F = c / λw = 343 / (6.0 × 10 <-3>) = 57.2 kHz
[0074]
Another factor that limits the effectiveness of high frequencies is that the sequence does not
function at (N-1) times the design frequency. That is, using the numbers of the previous example
also here: λ high = λ D / (N-1) λ D = 108.6 mm Therefore, λ high = 108.6 mm / 6 = 18.0 mm
Therefore, f high = 343 / λ D = 34 3 / 18.0 mm = 19 kHz
[0075]
In this example, the cutoff frequency determined by the design frequency is smaller than the
lower of the two limit frequencies, which is the actual high frequency cutoff point. Thus, the
lower of the two frequencies is the cutoff frequency, ie 19 kHz.
[0076]
FIG. 12 shows the top layer of the section element path of the small smartphone acoustic
diffusive manifold. This layer 1201 is formed with cuts 1202 that are sufficient, appropriate and
effective to accommodate the Cobra smartphone loudspeakers.
[0077]
Referring to Table 5, central element 1205 is provided with a path length of 16 mm.
02-05-2019
18
[0078]
The inlet 1207 to element "4" is adjacent to the central element 1205 but on the opposite side.
These elements are diverted to the lower layer via duct 1207 and reappear at the adjacent array
outlet at location 1203. These “4” paths are manipulated in length to be 78 mm long.
[0079]
On the same side as the central "0" element 1205, but on both sides a "1" element 1204 with a
path length of 31.5 mm is formed.
[0080]
FIG. 10 shows the lower layer 1301 of a miniature acoustic manifold.
This layer is also sufficient and effective to accommodate the Cobra smartphone loudspeaker via
recess 1302 and is co-located with upper layer 901.
[0081]
The upper layer 1201 supplies acoustic energy to the path 1304 towards the outlet through the
ducts 1207 and 1303, and transmits the acoustic energy to the outlet layer through the duct
1305.
[0082]
FIG. 15 shows the outlet dimensions of a standard smartphone manifold design.
The channel outlet is 2 mm wide and 1 mm high. Therefore, the cross-sectional area is 2 mm
<2>. Seven channels are spaced 3 mm apart. The output array is thus 20 mm wide and 1 mm
high.
02-05-2019
19
[0083]
The total area of the outlets is 7 × 2 mm <2> = 14 mm <2>.
[0084]
FIG. 16 shows the bifurcated area of the cobra diaphragm.
[0085]
As the cobra diaphragm has radiused corners, care must be taken to compensate for this in the
bifurcated region of the "2" element.
Since the area above the diaphragm is so small, it is not possible to provide a channel area of
normal size in the branched hard on collider area.
Thus, a compression scale is used.
[0086]
The diaphragm has a length of 12 mm and a height of 8 mm. Therefore, it has a cross-sectional
area of 12 × 8 = 96 mm <2>.
[0087]
Since the outlet is 14 mm <2>, this design includes a compression factor on the 96/14 = 6.9
scale. Thus, the volumetric velocity in the channel is 6.9 times the volumetric velocity at the
diaphragm. When implementing such high scale factors, care must be taken not to introduce
non-linear sound pressure levels into the channel.
[0088]
02-05-2019
20
FIG. 17 is a detailed view of the captured impulse response of the loudspeaker of the present
invention. This is an audio window from -20 cm before the maximum recording value to 20 cm
after the maximum recording value. The central part around t = 0 cm is shaped like a Gabor
wavelet. However, looking at the middle part, there are multiple signals in the measurement
before and after the Gabor wavelet. This may be "ringing" from a spectral box that has not been
sufficiently attenuated. FIG. 18 shows a Fast Fourier Transform FFT of the inventive manifold
developed for the Cobra manifold. The present embodiment is suitable for use in small household
appliances such as smartphones. Therefore, it has a function of temporarily marking an abrupt
phase change in the audio signal in the listening space (wavelet coding). Prior art criteria for FFT
show slight distortion due to the addition of this wavelet coding manifold. This spectral curve can
be equivalent by the host smartphone electronics.
[0089]
This system produces almost no spectral bass below 500 z. Converting the bass into equivalent
abrupt phase jumps at carrier frequencies above 500 Hz, so that the cold is perceived by the
brain via the temporal information channel rather than a spectral information channel requiring
FFT energy less than 500 Hz It seems plausible. Spectral energy below 500 Hz is simply not
physically supported by these small speaker drivers.
[0090]
The advantage of increasing the sound pressure level by increasing the volume velocity is to
increase the sound pressure level emitted to the listening space.
[0091]
FIG. 19 shows a manifold loudspeaker t01 emitting a sound field where phase anomalies
(temporal activity) occur at radius t02.
This phase transient is nominally the wavelet t03, in which there is a circular ring 104 with a
radius t02 around the manifold loudspeaker t01.
02-05-2019
21
[0092]
FIG. 20 shows the same manifold loudspeaker z01 emitting two phase anomalies causing two
temporal wavelet rings z06. A human standing in this radiated sound field listens to these
temporal rings z06 via both ears z03 and z02, producing a zero-phase image z05 in the human
perception system.
[0093]
FIG. 21 shows the stereo space of manifold loudspeakers y01 and y02 emitting temporal rings
y03 and y04 based on the phase anomalies of each channel. The monaural information in the
stereo mix exerts coherent acoustic energy along the center line of the speaker y06. The listener
y07 hears both direct energy from the manifold speakers y01 and y02. Also, experience the zerophase phantom sound field that is formed by the interaction between the left and right stereo
signals to produce a spectral sound field and a temporal sound field. A phase match exists in this
zero phase sound field. The slight difference between the left and right channels creates virtual
reality sound in the zero phase sound field.
[0094]
This image shows not only the depth of field but also the mirror image between the channels.
[0095]
FIG. 22 shows three manifold loudspeakers k01, k02, k03 arranged around the listener k07.
These three manifold loudspeakers k01, k02, k03 generate three direct sound fields from this
monaural content and three phantom zero-phase sound fields from the interaction sound field.
This provides a laterally immersive listening space.
[0096]
FIG. 23 shows the complete virtual reality audio space formed by five manifold loudspeakers
02-05-2019
22
y01, y02, y03, y04, y05. The manifold speakers y01, y02, y03, y04 are arranged in a fourchannel system horizontally around the listener y06. The manifold speaker y05 is disposed
above the listener y06. These manifold loudspeakers y01, y02, y03, y04, y05 produce five direct
zero-phase one-ear perceptions from the mono content from each sound source. These
loudspeakers generate the following six horizontal stereo zero-phase sound fields from the
interaction of horizontal sound sources:
[0097]
Then, four air zero-phase sound fields are formed from the following stereo interaction.
[0098]
This provides an immersive, realistic sound through the recording and manipulation of the five
channel audio signal.
Five channel coding of speech in digital files is known in the art. The zero-phase interval as
described above is a simulation of the "live" sound field. 6 horizontal stereo zero phase sound
fields 4 vertical zero phase sound fields 5 direct single zero phase sound fields
[0099]
FIG. 24 shows a manifold 2403 with a speaker driver 2404 rigidly mounted to a hard on collider.
In this dual configuration outline, the two parts are nested within one another. The outlet is
shaped to fit this structure into the dashboard of a car. FIG. 25 shows the manifold speaker
configurations 2405 and 2506 of FIG. 24 mounted within a car dashboard. The only visual
impact for the driver is the manifold outlet array. This is a conventional stereo configuration.
[0100]
FIG. 26 shows a manifold 2612 with a speaker driver 2613 rigidly mounted to a hard on collider.
In this dual configuration outline, the two parts are nested within one another. The outlet is
shaped to fit this structure to the back of a flat screen TV.
02-05-2019
23
[0101]
FIG. 27 is an isometric view of a flat screen TV 2716 with the structure of FIG. 26 mounted.
Outlets 2714 and 2715 are located on the front of the TV 2716.
[0102]
FIG. 28 is a rear isometric view of flat screen TV 2716 with manifolds 2714 and 2715 visible.
Because the manifolds 2714 and 2715 are made of plastic, they are injection molded as soon as
the entire back cover of the TV is installed. This significantly reduces the manufacturing cost.
[0103]
FIG. 29 shows a configuration "tone" consisting of a 500 Hz carrier. However, an abrupt 90
degree phase change occurs 21 every 3 milliseconds. As the Fast Fourier Transform shows, this
appears as a spectral combination of approximately 410 Hz and 750 Hz components. However,
at this tone, 333 Hz with 3 mil line spacing is dominant.
[0104]
FIG. 30 shows a configuration "tone" 24 consisting of an 800 Hz carrier. As the Fast Fourier
Transform shows, this appears to be spectrally only at 800 Hz 25.
[0105]
FIG. 31 shows a configuration "tone" 26 consisting of an 800 Hz carrier and small phase changes
(15 degrees) at 10 millisecond intervals. As the fast Fourier transform shows, this appears to be
spectrally only 800 Hz. However, a 100 Hz tone can be heard with a 10 millisecond phase
change. Smartphones are known to have small energy less than 500 Hz to 700 Hz. Physical
speaker drivers can not support tones below this area.
02-05-2019
24
[0106]
FIG. 32 shows a system in which a low frequency band is inserted into the pass band (700 Hz or
more) of a smartphone by first dividing an audio signal into less than 700 Hz components (3232)
and more than 700 Hz components (3229). The higher passing area 3229 is supplied to the
smartphone speaker 3231 and then passes through the phase modifier 3230. The lower portion
of the audio signal 3231 passes through a filter that extracts the bass information and is phase
shifted on the passband 3229 signal. In this method, the bass is encoded as a phase change into
an audio signal of 700 Hz or higher, and becomes perceptible as the bass passes through the
human temporal perception system.
[0107]
Similarly, other loudspeaker drivers suitable for dimensions and outputs of other consumer
electronics and industrial applications are also designed to improve the image of clarity, reach
and listening experience when coupled to the driver An acoustic diffusion manifold can be
provided.
[0108]
Other suitable sequences are those used in signal processing such as Barker codes, zero
autocorrelation sequences, or complementary sequences.
[0109]
The invention has been described with reference to specific embodiments.
It will be apparent to those skilled in the art that various modifications can be made and other
embodiments can be used without departing from the broader scope of the invention.
For example, other zero autocorrelation sequences or methods that achieve relative sequence
element time delays can also be used in the present invention. Thus, these and other variations of
the specific embodiments are also included in the present invention.
02-05-2019
25
Документ
Категория
Без категории
Просмотров
0
Размер файла
39 Кб
Теги
jp2018518880
1/--страниц
Пожаловаться на содержимое документа