close

Вход

Забыли?

вход по аккаунту

?

DESCRIPTION JPS616999

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 JPS616999
[0001]
TECHNICAL FIELD OF THE INVENTION The present invention relates to a method and apparatus
for imitating the transfer function of the outer ear when the outer ear is exposed to sound in a
free sound field, in other words, an electronic artificial head. [0002] In general, for example, a
partial system of a human outer ear model for simulating an ear signal that causes free sound
field sound propagation conditions when sound arrives from any direction, for example under
headphone playback conditions. Attempts to use an electroacoustic circuit for the alternative 9
are also known here. Also, it has been attempted to provide an electronic means that represents
the average outer ear transfer function appropriate for the assumed voice incident direction, and
a procedure that can be expressed as a kind of directional mixer unit. In order to construct and
implement such a directional mixer unit, the actual mechanism that generates the ear
communication characteristics is neglected, for example because it is unknown, and based on the
measurements performed on a plurality of subjects, The average outer ear transfer function may
be determined for a limited number of different voice incident directions. The external ear
transfer function thus determined for the individual voice incident directions allows switching of
the "directional mixer unit" to different directions, but has the following disadvantages. 1) With
regard to the ear transfer function, an averaging method that reliably provides an average
transfer characteristic that actually fits the information receiver configured by "human hearing"
has not been discovered today: 2) Limited Only a number of different voice incident directions
can be set; and 3) the cost and effort required for this task increase as the number of desired
configurable voice incident directions increases. Another way of imitating by electronic means a
so-called outer ear simulator that averages the average outer ear transfer function in the time
domain by 7 means, for example, appropriately averaging the transition responses measured for
all sound incident directions for the subject and It will be remembered. This method requires a
very large storage capacity according to the desired grid and turns. In this case, the output signal
11-04-2019
1
is considered to be a so-called composition of the input signal with two transition responses (left
and right ears) for the respective sound incident directions. However, such real-time signal
processing is practically impossible, since at least currently available signal processors can not be
implemented without a considerable burden of such signal processing. For the same reason, socalled Fourier transformation of the input signal with multiplication and inversion of the
corresponding transfer function has to be considered as impossible as well, and the 0
conventional mixer system has so-called panoramic control Since the individual microphone
signals can be distributed to the two channels of the stereo transmission system, the auditory
phenomenon between the two speakers when reproduced through the two speakers provided in
a normal stereo arrangement Spatial distribution occurs (localization).
However, this method has the following disadvantages. 1) The auditory phenomenon is present
only inside the spatial angle determined by the placement of the loudspeakers; 2) The height of
the auditory phenomena is often above the connection between the loudspeakers and depends
on the position of the listener relative to the loudspeakers And 3) during playback on
headphones, the aural phenomenon is in principle and (-6 ░ occurs in or on the listener's head,
as abnormal and / or unnatural ear signals are provided to the hearing) Problem requires a high
level of technology when reproducing the ear signal for a large number of sound incident
directions, so a practical useful implementation of the so-called electronic artificial scalp or outer
ear simulator is provided to date. It has not been. SUMMARY OF THE INVENTION Accordingly,
the object of the present invention is to open up new fronts in this field, and to consider the
transmission characteristics of the human outer ear completely, according to quantity and phase
for all frequencies and infinitely. The orientation can be realized without high cost and can
operate with particularly high accuracy despite the simple structure, ie, the transfer
characteristics of the electroacoustic circuit equivalent to the outer ear are free sound field sound
propagation conditions The present invention is to provide an outer ear simulator in the form of
a so-called electronic artificial head, which performs shimmyonion so as to correspond
approximately ideally to the transmission characteristics of the human outer ear. The invention
offers the advantage that an ear signal corresponding to any desired sound incident direction in
the free sound field can be generated, for example with little loss, for reproduction on
headphones. This makes it possible to realize a natural voice iR turn. Another advantage of the
present invention is that-despite the fact that the external ear transfer function has a very
complex structure, as is well known to the person skilled in the art, the external ear stain
according to the invention can be used The circuits required to implement the invention are
limited to at most simple filters, high-pass and low-pass filters, and resonant systems. The
parameters required to set the circuit during operation, such as the amount of time delay or the
cut-off frequency of the low pass filter, are physically set geometry using a model to analyze and
display the ear transfer function Directly from chemical dimensions. Aside from the time delays,
only a few of the filter meters of the model of the electronic artificial head represented by the
electronic circuit change for different sound incident directions. For this reason, the transfer
11-04-2019
2
function can be simulated by determining only the pairing and 2-3 parameters in the specific
sound incident direction.
Furthermore, it is also possible to add up the averaged geometrical property values to a fixed
gogram in order to mimic the mean ear transfer function, and thus in a further simplified
configuration which is particularly suitable for practical use. The corresponding control
parameters are calculated directly by the electronic logic element, microcomputer or
microprocessor and transferred to a controllable circuit block which shimmers the individual
elements of the electronic artificial head. By virtue of being able to realize any fine division of the
corner area in both the horizontal and median planes without requiring a large storage capacity,
it is possible to substantially reduce the free sound field including the vertical direction. It is
possible to generate an accurate ear signal for all speech incident directions. The electronic
artificial head according to the present invention can avoid localization of sound in the head and
generate an ear signal that is completely comfortable with the auditory sense so that any desired
direction of the auditory phenomenon can be adjusted. This will open up and break new ground
in the field of multitrack recording technology as well as six head technologies. In the field of
artificial head technology, the use of electronic artificial heads allows the signals of the
supporting microphones to be spatially correctly mixed into the head-related recordings. In
multi-track recording techniques, the use of an electronic artificial head converts the signals of
individual sound sources so that the entire range available to human hearing can be used to
generate auditory phenomena during head related playback. It is possible. You can change the
direction of the auditory phenomenon even during recording so that you can perform rough
movement of the movement of the sound source. Since the electronic artificial head and the
artificial head recording system both have or approximate the free sound field transfer
characteristics equivalent to each other, the speaker compatibility of the electronic artificial head
is comparable to the speaker compatibility of the artificial head recording system. In the present
invention, it is particularly advantageous to be able to change the fixedly stored geometric
characteristic values so that other ear transfer functions can later be simulated. Furthermore, the
personal transfer function of the subject or the effect of the hearing aid (i.e., the external ear
simulator all interface of the present invention can be effectively reproduced so as to effectively
reproduce the deformity of the IDO device, the device in-the-ear device, or the change in
tympanic impedance). The basic concept of the present invention will be described in detail
below with reference to the attached drawings, and the basic concept of the present invention is
that the physical predisposition of the ear transfer characteristic is the upper part of the human
body, It is to be subdivided by identification between predetermined simplified acoustic elements
such as the shoulder 1 head, the auricle including the auricular sinus, the ear canal and the
tympanic membrane and their reduction to the acoustic elements.
All these body parts have different influence on the ear transmission characteristics according to
11-04-2019
3
their specific geometrical dimensions and frequency, and the resulting ear transfer function is
the resonance 9 reflection and diffraction generated by all body parts The waves are composed
of complex and overlapping waves. The direction-dependent properties are substantially
determined by the elements constituted by the upper body part, the shoulders and the earlobe
edge. Basically, such a dependency is calculable (using K IRCI (diffraction of HOFF, @ min derived
from GREEN's theory), but such a calculation is intended by the present invention It is not
suitable for the explanation of the mean external ear transfer function and the graphical
representation required for that shime rayon in the model. Therefore, it is important to explain
the diffraction and reflection in one body part according to the system theory, without the need
of complicated calculation of complicated diffraction integral, which is embodied in the electronic
outer ear mimelator 'r technically It is a step that enables such an implementation by relatively
simple means, once successful. As is apparent from the above, the present invention does not
approach the problem from a layman's point of view, but rather determines complex rotational
and reflection conditions mathematically set, and the transfer function generated therefrom.
Starting from consideration and application, it proceeds by transferring the functions into a
(simplified) model which can be represented by electrical circuits by analytical observation. In
this case, for example, the earlobe or the head can be displayed by means of a specific circuit
based on a mathematical examination initially of the overlap of several diffractive objects. The
total outer ear transfer function is realized by complex combination of individual reflection /
diffraction sound components of the corresponding body part or body area that are to be
shimmed by the electric circuit block. The additional time delay allows any displacement of the
plane (the principle of superposition). In order to fully understand the present invention, it is
neither hopeful nor necessary to make it unnecessarily complicated by the explanation of
complex mathematical correlations throughout the present specification, but the meanings of the
above-mentioned matters will be exemplified below. explain. For example, the transfer function of
the auricular sinus, which forms the fundamental resonance, and which in this case has no
direction dependence as the influence of the impedance of the auricle and tympanic membrane,
It is sufficient to understand the auricular sinus as a stem having several openings connected
together and overlapping each other: this function is represented by the so-called Rayleigh
Stroop function represented by and by J1 And so-called Bessel functions.
Furthermore, the data corresponding to 1 n (data corresponding to the 'i length is limited, rn is
the data corresponding to the radius of n open 1) is limited, and K is limited to the wave number
(r4'c) k. Such a transfer function can be sufficiently approximated by a time delay in relation to a
resonant system as follows: in humans, vn is the open Lln amplification, Qn is the quality, and fO,
It is a resonant frequency. The parameters of the resonant system-resonant frequency 1 quality
and amplification-are functionally related to the geometrical dimensions of the pinna openingradius and depth. Thus, the present invention can be expressed mathematically with at least a
sufficient approximation of the correlation between the acoustically effective geometry outside
the living human being and the outer ear transmission function that can be measured. It is based
11-04-2019
4
on the recognition that there is a possible m-. Thus, based on the mean geometrical dimensions,
the mean outer ear transfer function can be determined in this way without additional effort for
all sound incident directions from 1 to 1. Because the physical correlation of the outer ear with
its transfer characteristics provides all the properties needed for the auditory signal analysis and
pattern recognition process, the average outer ear transfer function appropriately determines the
transfer properties required for human hearing. Show. Next, using the models based on known
systems in the field of communication technology (high band and low band filters, time delay
accumulators, resonant elements, etc.) on the physical predisposing factor of the external ear
transfer characteristic that can be expressed mathematically in this way On the assumption that
it can be approximated, the actual implementation is possible. Such a model can directly
approximate the outer ear transfer function by changing only a few parameters based on
physical property values. The system of the ear with direction-dependent transfer characteristics
represents one frequency-dependent distortion in terms of communication technology. The audio
signal is subject to frequency dependent distortion in response to the audio incident direction
while being converted into an ear signal for the information receiver, which is configured by
"human hearing". Next, taking a head which can be represented as a semi-elliptic body as an
example, a method of diffraction and reflection total approximation using a model will be
described. First, reference is made to FIG. 3 showing in a block diagram the directional part-1
channel only of the outer ear simulator 10. As shown in FIG. The three parts divided by the
alternate long and short dash lines represent approximations by circuit blocks of the head area
10a, the pinna area 10b, and the shoulder and upper body part 10c. For example, assuming that
the diffraction and reflection in the disk can be very well approximated by a simple model
consisting of two time delay elements, one low pass filter- Sound pressure curve at any one point
as a function of the voice incident direction by the circuit block (head region) 10a of the partial
model, for approximation of the head model, based on It is assumed that the approximation form
of the head and the ellipsoid of l- can be approximated.
The parameters of this model are functionally related to the geometrical dimensions of the
respective body. The parameters also take into account special cases such as sphere heads,
ellipses and discs. For example, in all circuit blocks shown in FIG. 3, K indicates a coefficient
element, T with index indicates a time delay element, TP indicates a low pass filter, and HP
indicates a high pass filter. The physical basis of the head model shown at 10a is the following: to
the block, and HP, directly incident sound waves the sound field reflected by the ellipsoid (head)
independently of the sound incident angle Add. The diffraction field generated by the action of
the edge is subdivided by a plurality of components. 3 to the element. The branch path including
all of BP3 and T2 represents a diffracted wave component in the direction toward the position of
the sound source, and the element 2 ░ BP2. A branch path including T and TPi represents a
component in a direction away from the position of the sound source. The cutoff frequency 2
time delay values and coefficients of the high-pass and low-pass filters are determined directly by
the size of the head, the direction of sound incidence and the position of the entrance of the ear
11-04-2019
5
canal. As indicated above for the head, the shoulders (3. ???? The shadow of the diffractive
object represented by Ps and TP, including the upper body portion (KOI HPO + TO and TPo) and
the pinnae (Ka, Ta and TP 8 and Kz, Tz), relates to the direction-dependent element shown in FIG.
Can be displayed with sufficient accuracy. The tables on the following pages contain geometrical
data of six subjects obtained by measurement. These data can be regarded as the geometric mean
value of the i-parameters of the individual approximate elements as shown in FIG. 3 and can be
considered as the basis of the design of the circuit elements. For imitation of the mean external
ear transfer function, the value of the mean geometric property value may be fully fixed
programmed, as described in more detail below. Table--Geometrical data (m = average, ?standard deviation) of 6 subjects (male) 1 Shoulder width (+ w +) 496 282 'Depth of waste (?)
269 2934 Shoulder slope (0) 23.3 2.94 * BP on top of the shoulder (Ball) 160 115 * BP from the
top (mm) 156 116 * BP from the bottom (?) 105 57 * Front surface with BP (mm) 116 684 Bp
from the back?) 102 59 * BP angle (') 11.9 4.610 * head width (r, 1 m) 177 1511 * head height
(theory) 61 1012 * head depth (wn) 218 613 * head radius (top) (Awards) 86 914 ?BP center
on top (?) 1115 * head radius (bottom) (distance) 66 1.31.6 * BP lower center (+ m) 25 71 head
radius (mm) 109 718 * BP upper center (biting) 41 1.319 * BP horizontal center (area) 14 1.020
* neck width (simplified) 104 821 * Neck depth (m) 117 1022 * neck angle (') 35.9 3.123 * BP
anterior chin (m) 94 524 * Ear height (theoretical) 625 * Ear width (?) 35 326 * Auricle tilt (')
12.4 5.327 * BP top center (Sho) 13 228 ?BP horizontal center (mm) 5 1291 Auricular Sinus
Height (Sophary) 30 230 ? ? Auricular Sinus Width (mm) 21 131 * Auricular Sinus Depth (111
111) 19 232 * Bl) Top Center (+11111) 4 133 * BP Lateral Center ( Cabinet 81: 34 ?head
radius, top (abbreviation) 81 11 or less, parameters used in the model of FIG. 3 (coefficient 1
time delay, high-pass filter, low-pass filter) - Although some mathematical relations are shown to
show what mathematical relationship is with the geometrical dimensions (parameters *) of the
tables in Table 1, the following formulas are given as examples only, and the present invention
Without limitation: Auricle: Ta (ra, rb, ?) = To (2 Ke), To = ra / cTz (rB + r1) + ?)-ToKe Kz (ra, rb,
?) -0,0 In the margin equation of 5 K (1 + 1 m (?) 1) (a (ra + rb + ?) = I Kz (ra, rb, ?) or less, ?
is the voice incident angle, and ra is half of the reference parameter 24 * And rb is half of the
reference parameter 25 *.
As another example, the head: 2?05 и K и ((1-31 n (of the door (r ?, a) / r ? + 10 0.5 и (1-cos
(?))-R / ( R + r ')] to K 3-05' и [(1 + 51 n (?) door (r '-R) / r' 0.5 10 0.5 и (1-cos (of) и R / (R + r '))
less than the margin ? '= Arc tan (R / r') ? ?? ?are C0BCr'7'r & 'C03 (?)) In the following
margin formula, R is ? of the value portion of the reference и ceramic 10 * of the table, Rv is a
part of the value of the reference parameter 7 *, Rh is a part of the value of the reference
parameter 8 *, Ro is a part of the value of the reference parameter 5 *, and shame is It is-of the
part of the value of the reference parameter 6 *. C is 6D at the speed of sound inside the ear, and
fg is the cut-off frequency of each high or low pass filter. In order to fully explain the structuredetermining properties, the influence on factors unrelated to the direction constructed by the ear
canal and the ear canal sinuses must be determined as well. The resonance characteristics of this
11-04-2019
6
auricular sinus can be very closely approximated by a bandpass system in the form of a series
resonant circuit. The parameters (resonance frequency 2 quality and amplification are also
functionally related to the geometrical dimensions of the tympanic membrane. The ear canal may
be understood as a tube with a complex terminal impedance, ie tympanic impedance. The system
can be sufficiently approximated by a model consisting of time delay elements, high pass filters
and coefficient elements. By considering the above, a very simplified block diagram of the outer
ear model shown in FIG. 1 is obtained. Only the left channel is shown in FIG. According to the
details described above and shown in FIG. 3, the individual blocks of this model are also shown in
FIG. 1 and are found in all living humans, and thus the extra-early transfer function of the ear
canal It represents the corresponding acoustic elements that limit the structure. As briefly
mentioned above, the model shown in FIG. 1 has a direction-dependent portion 12 for shimming
the directionality characteristics of the outer ear, and a portion independent of the direction for
shimming the free field external ear transfer function It is convenient to divide it into 13. During
playback via free field equalization headphones, an ear signal corresponding to the "average"
object's ear signal relative to the audio incident direction of the set can be generated via the free
field simulation output. The second output, shown as 14b in FIG. 1,-14a is a free field
equalization output-shimmyates the free field external ear transfer function.
The complete, schematically simplified model of FIG. 1 is the omitted form of the time delay
elements, the simple filter, the all pass filter and the adder only for the required circuit elements,
but Thus, the ear transfer function exhibits a very complicated structure in part. The parameters
of the circuit elements and blocks, such as the time delay value or the cut-off frequency of the
low pass filter, can be determined from the geometrically pre-measured geometrical dimensions,
ie from the above-listed tables. As a result, any desired voice input in the horizontal and median
planes, as would be immediately accepted by changing or modifying any of these predetermined
parameters or possibly any of the circuit blocks shown. A corresponding ear signal can be
generated with respect to the direction, such that such an electronic artificial hinge can have any
desired voice incident direction under free-field sound propagation conditions during playback
on headphones. A very serious conclusion is also obtained of providing a system capable of
generating representing ear signals and in particular capable of realizing natural and impressive
speech patterns. Similarly, in the case of playback on speakers, an improvement in transparency
similar to artificial head technology is achieved. This opens up new possibilities not only for the
special application to psychoacoustics discussed further below, but also for the new artistic
method of recording arrangements in the field of audio technology. In the model shown in FIG. 1
which includes one channel bandpass element, adder element, resonant element, etc. grouped by
circuit pack 1's, different time incident lights are considered, taking into consideration time delay
elements. Only a few filter parameters change with respect to direction. Therefore, by
determining these few parameters, the transfer function can be simulated for one predetermined
sound incident direction. In FIG. 1, the circuit block corresponding to the head region is denoted
by 10 A ?, the circuit block corresponding to the pinna and the pinna rim is denoted by 10 b ?,
11-04-2019
7
and the circuit block corresponding to debris and the upper body Is indicated by 100 '. The adder
elements which perform the additive superposition of the respective complex partial transfer
functions are denoted by reference numeral 15 in the figure. The circuit block of the portion
independent of the direction includes the ear canal and the auricular area, and is indicated by
reference numeral 16 in the figure. Another advantageous refinement of the invention is that all
time delays occurring in the external ear model are circuit block 1.
It is to be integrated into one basic time delay circuit block 17 connected in front of the OA ', 10'
and 10 'c and displaying and realizing the required signal deceleration and time delay. In this
regard, if the time delay is technically realized using an analog delay line, for example, in addition
to the poor ringing power / noise power ratio, frequency mixing products occur in the audio
frequency range In the present invention, the digital implementation of the time delay is
considered in order to obtain a high quality structure. Basically, this is achieved by placing all the
time delay elements associated with the individual part models or circuit element chains as
shown in FIG. 1, ie in front of the other individual circuits. Therefore, it is composed of a 16 bino)
ly'D converter operating at a sufficiently high rate, for example 44 kHz, to be employed. After
conversion, the quantized scan values are written to the soft register. Next, the delay time is
determined by the time difference between the read-in time and the read-in time of different
storage locations. This time difference is controlled by a microprocessor, described in more detail
below, which operates as a central controller of the individual elements. Due to the short access
time, all storage locations required for delay time simulation (eight delay times per channel-one
right and one left channel) are read out in one scan cycle. be able to. The scan value gold thus
delayed can then be output in time-division mode by means of a high speed D / A converter.
Based on this principle, it is only necessary to provide one or two D / A converters (one for each
channel). The filters and coefficients necessary for the simulation are preferably implemented
using a controllable operational amplifier. This is explained in more detail below. It is also
conceivable to realize the filter in digital form (for example by means of high-speed signals)
within the scope of the present invention, but it is desirable not to adopt this method at least at
the present time in terms of cost. As mentioned above, the electronic artificial head (outer ear
simulator) is preferably equipped with one central control unit i as shown in FIG. 1 because the
actual operation is very easy. For this purpose, a microprocessor 18 is provided, in which, for
example, the mean geometrical characteristic values required for the imitation of the mean
external ear transfer function are fixedly programmed.
In this case, the associated control parameters may be calculated accordingly by means of the
step 18 and transmitted directly to the controllable circuit block. According to this method, it is
possible to realize very fine division of the corner area in the horizontal plane and the median
plane without requiring a large storage capacity, so that the related ear can be used for all voice
incident directions in the free sound field. It can generate a signal. Furthermore, in this case, it is
11-04-2019
8
possible to change all fixed geometric characteristic values so that other ear transfer functions
can be imitated as well. The external ear shimmy of the present invention can also be coupled to
an external computer through an interface. This possibility is shown in detail in FIG. In the 19
position of FIG. 4, the key date of an external computer, such as a personal computer, is
associated with the microprocessor 18 '. Figures 2a and 2 show the quite surprising simulation
capabilities of the external ear simulator according to the invention. FIG. 2a is simulated
according to the invention--one free sound field outer ear transfer function (1) which does not
include a simulation of the upper body part in this example, and based on valid measurements, ie
determined experimentally Figure 2 shows the free sound field for the ear canal area (1) and the
shoulder and auricle edge area (2) and the auricular sinus (3), for example. Further illustrated are
simulations of individual acoustically valid parameters that can be expressed as partial ear
transfer functions. In this case, the two partial curves constitute the outer ear transfer function
(1) e of FIG. As mentioned above, the individual circuit elements of FIG. 3 represent, in more
detail, the head area, the pinnae area and the shoulder / upper body area of the circuit block of
FIG. Following the basic time delay circuit block 17 realized by the use of the individual adder
elements (not listed earlier) together with a final adder element 15 'having an output connection
20 leading to an element independent of direction 15a, 15b, 15c, 15d). The circuit elements of
FIG. 3 constitute the analog part of the micro-poloprocessor control outer ear simulator for the
realization of coefficient elements, high-pass and low-pass filters and adder elements for
combining their output signals. The detailed example shown in FIG. 4 is a schematic drawing
showing one possible implementation of the external ear simulator according to the invention.
The outer ear simulator comprises a block 21 which includes operations and human power
elements, an indicator and is assigned to the microprocessor system 18 '. Microprocessor 18
'further includes or cooperates with central timing controller 22. The microprocessor acts on the
parameters of the analog circuit channel 24 via a plurality of connection lines 23a, 23bt. In this
embodiment, eight analog circuit channels are provided, whose output terminals are connected
to the adder element 15 ''. Depending on the nature and structure of the model, the analog circuit
channel 24 may include first and / or third order low pass and high pass filters 24a, 24b, a band
pass filter 24c, and an amplification factor of -1 ... +1 And the so-called coefficient element 24d.
The time delay elements of the individual channels are realized as digital delay lines and for this
purpose only one A / D conversion has to be carried out in the digitizing block 26 connected
after the human low pass filter 25 Arranged as. After conversion, the quantized scan values are
written to freely addressable storage 27 (RAM delay line storage). The delay time obtained
between the writing and reading of the scan values at different storage locations determines the
time difference, and the length of the Renosta is determined by the maximum required delay
time. Considering that the storage access is very fast compared to the previously mentioned
constant strain rate of preferably 44 kHz, all scan values needed to simulate different delay times
are taken of one scan cycle. It can be read out continuously. Thus, with the correspondingly high
speed D / A converter 26, the scan values thus obtained for different time delays can be
11-04-2019
9
reconverted in time division mode. To that end, a time division multi-zone switch 29 is provided
following the signal recovery block 28, which is controlled by the central timing controller and
whose output terminals are respectively connected to different analog circuit channels 24. The
combination of analog and digital circuit elements offers the possibility to realize such a system
without any problems, and at the same time guarantees a very flexible high quality system for
simulating the ear transfer function. Further, 30m in FIG. 4 is, for example, one right channel
portion, and 30b is an associated left channel portion. The low pass filters 31a and 31b are
respectively connected after the adder element 15 ', the output terminal 32a of the low pass filter
31a supplies the right ear signal, and the output 2 terminal 32b of the low pass filter 31b
supplies the left ear signal Do.
FIG. 5 shows a possible implementation of the circuit element designed as a first order low pass
or high pass filter as desired. The filter is constituted by a so-called "operational
transconductance amplifier-TA) 33 configured as a controllable resistor, the forward
transconductance of which is the inverse value of the amplification, and the externally supplied
DC current Ilt It can be adjusted. According to the supply point of such a DC input signal, the
transfer function obtained for the whole configuration is either a low pass filter transfer function
or a high pass filter transfer function 1, and finally, a normal operational amplifier 34 after 0 TA
33 Is connected. The control current is provided from the lower circuit portion, and the control
voltage Ust is supplied to the operational amplifier 35 and is also supplied to 0TA 33 via the FF
and T transistor 36. Other essential components are only the capacitor C connected to the feed
puncture path, and the resistors R3 and R4 provided on the input wiring portion in a reverse
connection and connected to the feedback line 37. The cut-off frequency proportional to the
direct current control current 1st is obtained in this type of circuit, for example, from the
following equation: The circuit shown in FIG. 5 generally provides a voltage controlled low / high
pass filter element Do. FIG. 6 is a block diagram of the interface circuit that generates the control
voltage Uste that can be tapped at the output terminal 38 of the circuit and is required to set the
parameters of the filter and coefficient elements. The microprocessor 18 '(FIG. 4) writes
parameter data words into the data register 40 via the data bus 39f. At the output of the data
register, the data word is converted into a voltage, for example 0... -10 V, by means of a digital /
analog converter 41 to which a current / voltage converter 42 is later connected. Then, the
channel of the analog multiplexer 4401 connected to the output terminal of the current / voltage
converter 42 is addressed via the address register 43 operated by the same data bus, and
thereby generated. The supply voltage is supplied to the corresponding output storage circuit
(sample ten hold). A separate S + f (circuit 45 is provided for each filter element to be controlled.
The sample / hold circuit 45 comprises only a storage capacitor C and a voltage follower 46 of
very high resistance as an operational amplifier. When the capacitor C is charged, the channel is
blocked by the inhibit signal from the address register 43.
11-04-2019
10
The entire process is performed periodically in all other channels as well. Thus, the voltage
appearing on the holding capacitor C is constantly updated. Additional decode logic 47 generates
all of the charge pulses for the two registers 43 and 40 by means of the address bus input line
48 and the control bus input line 49 of the microprocessor system 18 '. Finally, FIGS. 7 and 8 are
graphs showing the freefield external ear transfer function for two different directions (0 ░ and
2 ░) in the free sound field (horizontal plane) of the left ear of a living subject It is. The solid line
represents the curve obtained by calculation according to the subject matter (model) of the
present invention, and the two dotted lines above and below the solid line represent the standard
deviation of six measurements performed on the same subject. It can be seen that the present
invention achieves its object, namely the realization of the electronic artificial head with very
high precision. Particularly suitable applications of the electronic artificial head according to the
invention which can be used instead of the natural outer ear for the purpose of converting
speech and ear signals are, inter alia, three areas: l) Psychoacoustics research In order to easily
realize a special outer ear transfer function, for example, in the field of hearing aids, it is possible
to change the individual parameters simply and quite quickly, for example the influence of
hearing aids. Alternatively, the effect of the one or both ear hearing aid can be easily simulated L
/-. 2) In the field of medical diagnostics, it is used to test the degree of hearing of the voice in a
directional hearing or noisy environment. A device according to the invention, for example a
listening test. Directional listening tests, especially under free sound field conditions, can be
carried out without the need to provide a room with low opposition and without having to invest
considerable cost and effort. 3) In the field of sound engineering, it is used for synthetic
generation of head related recordings, and it is possible to mix the signal for arbitrary voice
incident direction. For example, artificial head recording. All features described or shown in the
above description, the appended claims and the drawings are considered essential to the
invention either alone or in any desired combination.
[0002]
Brief description of the drawings
[0003]
FIG. 1 is a very schematic block diagram of the external ear simulator according to the invention,
which also shows the division of direction-dependent circuit elements and circuit elements which
are not related to the direction, FIGS. 2a and 2b 6 is a curve representing the contrast of the free
field external ear transfer function (1) simulated according to the invention with the measured
function (n) in relation to frequency.
11-04-2019
11
And the individual acoustically effective parameters, ie, the simulation of the ear canal (curve 1),
the edge formed by the shoulders and the pinna (curve 2) and the auricle (curve 3) Graph
including curves (1), (2) and (3), FIG. 3 is a diagram showing a detailed example limited to one
channel direction dependent element in a simplified form (outer ear simulator one direction
Single-channel block diagram of the active part), Figure 4 is suitable for practical use where the
nosolameters of the individual circuit elements are controlled by the micro zolosser system using
stored mean geometrical property values. A schematic diagram showing one embodiment of the
present invention, FIG. 5 shows in detail one possible embodiment of a voltage controlled low /
high pass filter of the kind that can be used to realize the electronic artificial head according to
the present invention Fig. 6 shows a micro zoro In accordance with the invention, a block
diagram showing an interface circuit for generating a control voltage to change each no
parameter for individual circuit blocks under control of a sensor, and FIGS. A function of the
function calculated using the model to be realized, and a solid curve representing the function,
and a dotted curve representing the standard deviation of a predetermined number of
measurements performed on the same object. FIG. 7 is a graph showing the free sound field
outer ear transfer function (here for the left ear) for two different sound incident angles in the
free sound field (horizontal plane). 10 outer ear simulator 108 'head area circuit block 10b'
earlobe and auricle edge area circuit block 10c 'scrap and upper body area circuit block 12
direction dependent portion 13 и и и Direction-independent part, 15 ? и и и Adder element, 15a-15d
и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и и Circuit clock, 8 '... Micro
cessor, 22... Central timing control device 24 .. 24 analog circuit channel, 24 a .. low pass filter,
24 b .. high pass filter, 24 c .. band -Y illk, 24 d иии coefficient element, 26 и digitizing block, 27 и и
storage unit, 29 и и и time division multiplex changeover switch. Change the float of the margin
drawing below to the contents L)
11-04-2019
12
Документ
Категория
Без категории
Просмотров
0
Размер файла
32 Кб
Теги
description, jps616999
1/--страниц
Пожаловаться на содержимое документа