CN106412793B - The sparse modeling method and system of head-position difficult labor based on spheric harmonic function - Google Patents
The sparse modeling method and system of head-position difficult labor based on spheric harmonic function Download PDFInfo
- Publication number
- CN106412793B CN106412793B CN201610802607.0A CN201610802607A CN106412793B CN 106412793 B CN106412793 B CN 106412793B CN 201610802607 A CN201610802607 A CN 201610802607A CN 106412793 B CN106412793 B CN 106412793B
- Authority
- CN
- China
- Prior art keywords
- head
- represent
- difficult labor
- minimum phase
- position difficult
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000005259 measurement Methods 0.000 claims abstract description 60
- 238000001228 spectrum Methods 0.000 claims abstract description 39
- 208000037805 labour Diseases 0.000 claims description 165
- 240000006409 Acacia auriculiformis Species 0.000 claims description 31
- 210000005069 ears Anatomy 0.000 claims description 31
- 238000012546 transfer Methods 0.000 claims description 18
- 238000011156 evaluation Methods 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 12
- 238000002790 cross-validation Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 description 76
- 210000003128 head Anatomy 0.000 description 28
- 230000015654 memory Effects 0.000 description 10
- 230000008569 process Effects 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000007480 spreading Effects 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 210000003454 tympanic membrane Anatomy 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04S—STEREOPHONIC SYSTEMS
- H04S7/00—Indicating arrangements; Control arrangements, e.g. balance control
- H04S7/30—Control circuits for electronic adaptation of the sound field
- H04S7/305—Electronic adaptation of stereophonic audio signals to reverberation of the listening space
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04S—STEREOPHONIC SYSTEMS
- H04S2400/00—Details of stereophonic systems covered by H04S but not provided for in its groups
- H04S2400/05—Generation or adaptation of centre channel in multi-channel audio systems
Abstract
The invention discloses the sparse modeling methods and system of a kind of head-position difficult labor based on spheric harmonic function.Wherein, this method includes：Head-position difficult labor to be measured is handled, minimum phase head-position difficult labor is generated, and remove the minimum phase amplitude equalizing value of full measurement direction, obtains Spatial Difference minimum phase head-position difficult labor；Spatial Difference minimum phase head-position difficult labor is modeled, obtains sparse spherical harmonic coefficient；By sparse spherical harmonic coefficient interpolation, and according to the continuous head-position difficult labor of the orientation in the space generation total space.The interpolation spectrum distortion degree of the continuous HRTF of the total space is reduced as a result, reduces computation complexity, required amount of storage is small, convenient for being applied to progress dynamic environment drafting in practical virtual reality system, is not limited by individual individual character, with high robust, convenient for application in the actual environment.
Description
Technical field
The present embodiments relate to signal processing technology fields, and in particular to a kind of head associated transport based on spheric harmonic function
The sparse modeling method and system of function.
Background technology
The outburst of field of virtual reality causes virtual auditory to receive more and more attention.Virtual reality includes virtual vision
And virtual auditory, wherein, the Important Problems of virtual auditory technology are to restore the location feature identical with the natural sense of hearing.The mankind's listens
Feel process may be generally viewed as sound source-channel-reception model, and wherein channel includes sound source and passes through the diffraction of human body different parts, does
It disturbs, eventually arrives at the process of eardrum, a space digital filter, referred to as head-position difficult labor (Head- can be regarded as
Related Transfer Function, HRTF), it contains all spectrums caused by interaction between sound wave and body part
Feature.Since everyone physiological structure is not quite similar, HRTF spectrum signatures are extremely personalized.However, it is difficult to per each and every one
Body measures HRTF in the total space.It is difficult that the HRTF databases of intensive measurement are effectively stored that another problem, which is,.It is in this regard, a kind of
Solution is to model HRTF to lower dimensional space, will be empty such as using principal component analytical method or spatial principal component analysis
Between variation be modeled as the joint of a small amount of principal component.However, these methods are difficult that the HRTF of discrete measurement is interpolated to the total space
Continuous HRTF.Another method is using modeling (the Spherical Harmonics-based based on surface spheric harmonic function
Modeling, SHM)) method, its major advantage be HRTF can the total space be modeled as relatively small amount ball it is humorous extension system
Several linear combination, therefore, entire HRTF databases only need low volume data to represent.
Spheric harmonic function is blocked using degree, spend play the role of in the modeling based on spheric harmonic function it is very important.Degree
Selection affect the complexity of model and spectrum distortion performance.Directly selected in conventional method a number bigger than normal or according to
The subjective sense of hearing selection of people.However, there are problems that two：
(1) due to the importance of subband different in the Auditory Perception of people be it is different, to all subbands distribute phase
Same degree is unreasonable, and some subbands set the research of the different number of degrees not have too many theoretical foundation；
(2) if the number of degrees of selection are less than normal, it will model poor fitting is caused, so as in interpolation HRTF continuous for the total space
Larger spectrum distortion will be generated with high probability.
In view of this, it is special to propose the present invention.
Invention content
Above-mentioned one or more to solve the problems, such as, the present invention provides a kind of head associated transport letters based on spheric harmonic function
Several sparse modeling methods can simply obtain the continuous head-position difficult labor of the total space of low spectrum distortion.It is in addition, of the invention
Embodiment also provides a kind of sparse modeling of the head-position difficult labor based on spheric harmonic function.
To achieve these goals, according to an aspect of the invention, there is provided following technical scheme：
A kind of sparse modeling method of the head-position difficult labor based on spheric harmonic function, the method include at least：
Obtain the head-position difficult labor to be measured；
The head-position difficult labor to be measured is handled, generates minimum phase head-position difficult labor, and removes complete survey
The minimum phase amplitude equalizing value in direction is measured, obtains Spatial Difference minimum phase head-position difficult labor；
The Spatial Difference minimum phase head-position difficult labor is modeled, obtains sparse spherical harmonic coefficient；
By the sparse spherical harmonic coefficient interpolation, and according to the continuous head associated transport letter of the orientation in the space generation total space
Number.
Further, the processing head-position difficult labor to be measured, generates minimum phase head associated transport letter
Number, and the minimum phase amplitude equalizing value of full measurement direction is removed, Spatial Difference minimum phase head-position difficult labor is obtained, is had
Body includes：
Non-minimum phase bit position in the head-position difficult labor to be measured is removed, obtains minimum phase head associated transport
Function；
The minimum phase amplitude equalizing value of full measurement direction is calculated according to the following formula, obtains the general character component of directional correlation：
Wherein, the djRepresent the horizontal angle at j-th of position and the elevation angle；The S represents to measure the total number of position；Institute
State | Hmin(dj,fi) | represent the amplitude of the minimum phase head-position difficult labor；The fiRepresent i-th of frequency band；The i takes
Positive integer；The Havg(fi) represent the general character component of the directional correlation；
According to the following formula in the measurement orientation of each head-position difficult labor, removed from minimum phase log-magnitude
The general character component of the directional correlation obtains Spatial Difference minimum phase head-position difficult labor：
Hp(ds,fi)=20log10|Hmin(ds,fi)|-Havg(fi),
Wherein, the dsRepresent the horizontal angle at s-th of position and the elevation angle；It is described | Hmin(ds,fi) | represent the minimum
The amplitude of phase head-position difficult labor；The Havg(fi) represent the general character component of the directional correlation；The Hp(ds,fi) table
Show s-th of position, i-th of frequency band fiSpatial Difference minimum phase amplitude.
Further, it is described that the Spatial Difference minimum phase head-position difficult labor is modeled, it obtains sparse
Spherical harmonic coefficient specifically includes：
Spheric harmonic function at setting measurement position is calculated according to the following formula：
Wherein, the l represents the number of degrees of the spheric harmonic function；The m represents the exponent number of the spheric harmonic function；The n tables
Show the number of degrees of Legendre function；It is describedRepresent the Legendre function that the number of degrees are n, the exponent number is m；Institute
It statesRepresent the measurement position for the spheric harmonic function at d；The d=(θ, φ) measures position d, wherein water shown in representing
The straight angle is θ, elevation angle φ；
To the square error of the Spatial Difference minimum phase head-position difficult labor amplitude and its spheric harmonic function
It is regular to carry out L1, the model minimized the error；
The optimal degree of rarefication of the model is obtained by K cross validation method, so as to obtain sparse spherical harmonic coefficient.
Further, it is described by the sparse spherical harmonic coefficient interpolation, and according to the company of the orientation in the space generation total space
Continuous head-position difficult labor, specifically includes：
The continuous head correlation of minimum phase measured on position is rebuild by the sparse spherical harmonic coefficient according to the following formula to pass
The time difference that defeated function amplitude estimation reaches left ear and auris dextra with sound wave is estimated：
Wherein, it is describedRepresent the measurement position dsThe left and right ear Spatial Difference minimum phase for locating i-th of frequency band connects
The estimation of continuous head-position difficult labor amplitude；It is describedRepresent l from setMiddle value, l=0 ..., Nm；It is describedRepresent the
The N of i frequency bandmIt is not the position where 0 coefficient in a spherical harmonic coefficient；It is describedRepresent the measurement position dsLocate sound wave
The time difference for reaching left ear and auris dextra is estimated；It is describedRepresent the sparse spherical harmonic coefficient；The Yl(ds) represent the measurement
Position dsThe spheric harmonic function at place；L takes positive integer；
The continuous head-position difficult labor of left and right ear is determined according to the following formula：
Wherein, it is describedRepresent the left ear head-position difficult labor；It is describedRepresent the auris dextra head
Related transfer function；The T0Represent that sound wave reaches the time of auris dextra, T0=Lr/ v, wherein, the LrRepresent sound wave apart from auris dextra
Distance, the v represents the velocity of sound；
According to the different orientation in space, determine the continuous head-position difficult labor of left and right ear, obtain the total space
Continuous head-position difficult labor.
Further, the method further includes：
Logarithmic spectrum distortion evaluation is carried out to the continuous head-position difficult labor of the total space according to the following formula：
Wherein, it measures and counts out in the S representation spaces, the NfRepresent number of sub-bands；It is describedDescribed in expression
The left and right continuous head-position difficult labor of ear；H (d, the fk) represent to measure the obtained left and right ear head-position difficult labor；
The fkRepresent k-th of frequency band；The k1With the k2The frequency range for representing comparison respectively is from kth1A frequency band is to kth2It is a
Frequency band；It is describedRepresent the spatial position set of all measurement head-position difficult labors；
Absolute log spectrum distortion is carried out to the continuous head-position difficult labor of the total space and opposite logarithm spectrum distortion is commented
Estimate.
To achieve these goals, according to another aspect of the present invention, a kind of head based on spheric harmonic function is additionally provided
The sparse modeling of related transfer function, the system include at least：
A kind of sparse modeling of the head-position difficult labor based on spheric harmonic function, the system include at least：
Acquisition module, for obtaining the head-position difficult labor to be measured；
Processing module is connected with the acquisition module, and for handling the head-position difficult labor to be measured, generation is most
Small phase head-position difficult labor, and the minimum phase amplitude equalizing value of full measurement direction is removed, obtain Spatial Difference minimum phase
Potential head related transfer function；
Modeling module is connected with the processing module, for the Spatial Difference minimum phase head associated transport letter
Number is modeled, and obtains sparse spherical harmonic coefficient；
Generation module is connected with the modeling module and is used for through the sparse spherical harmonic coefficient interpolation, and according to space
Orientation generates the continuous head-position difficult labor of the total space.
Further, the processing module specifically includes：
First removal module：For removing non-minimum phase bit position in the head-position difficult labor to be measured, obtain
Minimum phase head-position difficult labor；
The general character component acquisition module of directional correlation is connected with the described first removal module, based on according to the following formula
The minimum phase amplitude equalizing value of full measurement direction is calculated, obtains the general character component of directional correlation：
Wherein, the djRepresent the horizontal angle at j-th of position and the elevation angle；The S represents to measure the total number of position；Institute
State | Hmin(dj,fi) | represent the amplitude of the minimum phase head-position difficult labor；The fiRepresent i-th of frequency band；The i takes
Positive integer；The Havg(fi) represent the general character component of the directional correlation；
Second removal module, is connected with the general character component acquisition module of the directional correlation, for being existed according to the following formula
In the measurement orientation of each head-position difficult labor, the general character point of the directional correlation is removed from minimum phase log-magnitude
Amount, obtains Spatial Difference minimum phase head-position difficult labor：
Hp(ds,fi)=20log10|Hmin(ds,fi)|-Havg(fi),
Wherein, the dsRepresent the horizontal angle at s-th of position and the elevation angle；It is described | Hmin(ds,fi) | represent the minimum
The amplitude of phase head-position difficult labor；The Havg(fi) represent the general character component of the directional correlation；The Hp(ds,fi) table
Show s-th of position, i-th of frequency band fiSpatial Difference minimum phase amplitude.
Further, the modeling module specifically includes：
Computing module, for calculating the spheric harmonic function at setting measurement position according to the following formula：
Wherein, the l represents the number of degrees of the spheric harmonic function；The m represents the exponent number of the spheric harmonic function；The n tables
Show the number of degrees of Legendre function；It is describedRepresent the Legendre function that the number of degrees are n, the exponent number is m；Institute
It statesRepresent the measurement position for the spheric harmonic function at d；The d=(θ, φ) measures position d, wherein water shown in representing
The straight angle is θ, elevation angle φ；
Regular module is connected with the computing module, for the Spatial Difference minimum phase head associated transport letter
Number amplitudes with its described spheric harmonic function square error progress L1 it is regular, the model minimized the error；
Degree of rarefication acquisition module is connected with the regular module, and the model is obtained for passing through K cross validation method
Optimal degree of rarefication, so as to obtain sparse spherical harmonic coefficient.
Further, the generation module specifically includes：
Module is rebuild, for rebuilding the minimum phase measured on position by the sparse spherical harmonic coefficient according to the following formula
The time difference that continuous head-position difficult labor amplitude Estimation reaches left ear and auris dextra with sound wave is estimated：
Wherein, it is describedRepresent the measurement position dsThe left and right ear Spatial Difference minimum phase for locating i-th of frequency band connects
The estimation of continuous head-position difficult labor amplitude；It is describedRepresent l from setMiddle value, l=0 ..., Nm；It is describedRepresent the
The N of i frequency bandmIt is not the position where 0 coefficient in a spherical harmonic coefficient；It is describedRepresent the measurement position dsLocate sound wave
The time difference for reaching left ear and auris dextra is estimated；It is describedRepresent the sparse spherical harmonic coefficient；The Yl(ds) represent the measurement
Position dsThe spheric harmonic function at place；L takes positive integer；
First determining module is connected with the reconstruction module, for determining that the continuous head of left and right ear is related according to the following formula
Transfer function：
Wherein, it is describedRepresent the left ear head-position difficult labor；It is describedRepresent the auris dextra head
Related transfer function；The T0Represent that sound wave reaches the time of auris dextra, T0=Lr/ v, wherein, the LrRepresent sound wave apart from auris dextra
Distance, the v represents the velocity of sound；
Second determining module is connected with first determining module, for according to the different orientation in space, determining described
The left and right continuous head-position difficult labor of ear, obtains the continuous head-position difficult labor of the total space.
Further, the system also includes：
Logarithmic spectrum distortion evaluation module is connected with the generation module, for according to the following formula to the continuous of the total space
Head-position difficult labor carries out logarithmic spectrum distortion evaluation：
Wherein, it measures and counts out in the S representation spaces, the NfRepresent number of sub-bands；It is describedDescribed in expression
The left and right continuous head-position difficult labor of ear；H (d, the fk) represent to measure the obtained left and right ear head-position difficult labor；
The fkRepresent k-th of frequency band；The k1With the k2The frequency range for representing comparison respectively is from kth1A frequency band is to kth2It is a
Frequency band；It is describedRepresent the spatial position set of all measurement head-position difficult labors；Logarithmic spectrum distortion evaluation module, with institute
It states generation module to be connected, for carrying out absolute log spectrum distortion and relatively right to the continuous head-position difficult labor of the total space
Number spectrum distortion assessment.
It can be seen from the above technical proposal that the sparse modeling side of the head-position difficult labor the present invention is based on spheric harmonic function
Method and system have the advantages that：
Head-position difficult labor to be measured is handled, generates minimum phase head-position difficult labor, and is removed complete
The minimum phase amplitude equalizing value of measurement direction obtains Spatial Difference minimum phase head-position difficult labor；To Spatial Difference
Minimum phase head-position difficult labor is modeled, and obtains sparse spherical harmonic coefficient；By sparse spherical harmonic coefficient interpolation, and according to sky
Between orientation generation the total space continuous head-position difficult labor.It reduces as a result, dry in the head-position difficult labor of measurement
It disturbs, remains the key feature of head-position difficult labor to the full extent, thus greatly reduce inserting for the continuous HRTF of the total space
It is worth spectrum distortion degree；And computation complexity is also reduced, required amount of storage is small, convenient for being applied to practical virtual reality system
Dynamic environment drafting is carried out in system；In addition it can carry out number of degrees selection automatically with measurement object and data, therefore not by individual
The limitation of individual character has high robust, convenient for application in the actual environment.
Other features and advantages of the present invention will be illustrated in the following description, also, partly becomes from specification
It obtains it is clear that being understood by implementing the present invention.Objectives and other advantages of the present invention can be by the explanation write
Specifically noted method is realized and is obtained in book, claims and attached drawing.
Description of the drawings
A part of the attached drawing as the present invention, for providing further understanding of the invention, of the invention is schematic
Embodiment and its explanation are for explaining the present invention, but do not form inappropriate limitation of the present invention.Obviously, the accompanying drawings in the following description
Only some embodiments, to those skilled in the art, without creative efforts, can be with
Other accompanying drawings are obtained according to these attached drawings.In the accompanying drawings：
Fig. 1 is the sparse modeling side of the head-position difficult labor based on spheric harmonic function according to an exemplary embodiment
The flow diagram of method；
Fig. 2 is the sparse modeling of the head-position difficult labor based on spheric harmonic function according to another exemplary embodiment
The structure diagram of system.
These attached drawings and word description are not intended to the protection domain limiting the invention in any way, but by reference to
Specific embodiment illustrates idea of the invention for those skilled in the art.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with specific embodiment, and reference
Attached drawing, the present invention is described in more detail.
It should be noted that in attached drawing or specification description, similar or identical part all uses identical figure number.And
In the accompanying drawings, with simplify or facilitate mark.Furthermore the realization method for not being painted or describing in attached drawing is technical field
In form known to a person of ordinary skill in the art.In addition, though the demonstration of the parameter comprising particular value can be provided herein, it is to be understood that
Parameter can be similar to be worth accordingly in acceptable error margin or design constraint without being definitely equal to corresponding value.
The basic thought of exemplary embodiment of the present is the rarefaction model for building head-position difficult labor, by mould
Type error progress L1 is regular, obtains representing the humorous spreading coefficient of ball of big data quantity head-position difficult labor, so as to which head correlation be passed
Defeated function is described using one group of vectorization parameter, finally can obtain the total space from the head-position difficult labor of discrete measurement
The continuous head-position difficult labor in direction, the virtual auditory drafting for dynamic scene provide basis.
The embodiment of the present invention provides a kind of sparse modeling method of the head-position difficult labor based on spheric harmonic function.Such as Fig. 1
Shown, this method can include：Step S100 to step S130.
S100：Obtain head-position difficult labor to be measured.
S110：Head-position difficult labor to be measured is handled, generates minimum phase head-position difficult labor, and removes complete survey
The minimum phase amplitude equalizing value in direction is measured, obtains Spatial Difference minimum phase head-position difficult labor.
The Spatial Difference minimum phase head-position difficult labor obtained in this step be without direction general character ingredient most
Small phase head-position difficult labor, wherein, the transmission feature of the direction is mainly included in the head-position difficult labor in each direction.
S120：Spatial Difference minimum phase head-position difficult labor is modeled, obtains sparse spherical harmonic coefficient.
Specifically, this step calculates the sparse humorous system of ball using Spatial Difference minimum phase head-position difficult labor amplitude
Number.The head-position difficult labor of big data quantity is carried out rarefaction representation by this step.
S130：By sparse spherical harmonic coefficient interpolation, and according to the continuous head associated transport of the orientation in the space generation total space
Function.
The embodiment of the present invention automatically removes and contributes modeling smaller coefficient, in the head-position difficult labor for reducing measurement
Interference, remain the key feature of head-position difficult labor to the full extent, thus greatly reduce the continuous head phase of the total space
The interpolation spectrum distortion degree of transfer function is closed, in addition, the embodiment of the present invention can carry out the number of degrees automatically with measurement object and data
The algorithm of selection, therefore do not limited by individual individual character, there is higher robustness.
In some embodiments, step S110 may further include：Step S111 to step S113.
S111：Non-minimum phase bit position in head-position difficult labor to be measured is removed, obtains minimum phase head correlation biography
Defeated function.
In this step, the amplitude and phase of minimum phase head-position difficult labor can be determined according to the following formula：
|Hmin(θ, φ, f) |=| H (θ, φ, f) |
Wherein, H (θ, φ, f) represents head-position difficult labor to be measured；θ represents to measure the horizontal angle of position；φ is represented
Measure the elevation angle of position；F represents frequency；ξ represents the instantaneous frequency at a certain moment.
S112：The minimum phase amplitude equalizing value of full measurement direction is calculated according to the following formula, obtains the general character of directional correlation
Component：
Wherein, djRepresent the horizontal angle at j-th of position and the elevation angle, dj=(θj,φj)；S represents to measure the sum of position
Mesh；|Hmin(dj,fi) | represent the amplitude of minimum phase head-position difficult labor；fiRepresent i-th of frequency band；I takes positive integer；Havg
(fi) represent directional correlation general character component (the minimum phase amplitude equalizing value of i.e. full measurement direction).
S113：According to the following formula in the measurement orientation of each head-position difficult labor, from minimum phase log-magnitude
The general character component of middle removal directional correlation, obtains Spatial Difference minimum phase head-position difficult labor：
Hp(ds,fi)=20log10|Hmin(ds,fi)|-Havg(fi),
Wherein, dsRepresent the horizontal angle at s-th of position and the elevation angle, ds=(θs,φs)；|Hmin(ds,fi) | represent minimum
The amplitude of phase head-position difficult labor；Havg(fi) represent directional correlation general character component；Hp(ds,fi) s-th of position of expression,
I-th of frequency band fiSpatial Difference minimum phase amplitude.
In some embodiments, step S120 may further include：Step S121 to step S123.
S121：Spheric harmonic function at setting measurement position is calculated according to the following formula：
Wherein, l represents the number of degrees of spheric harmonic function；M represents the exponent number of spheric harmonic function；N represents the number of degrees of Legendre function；The expression number of degrees are n, the Legendre function that exponent number is m；It represents to measure position as the spheric harmonic function at d (i.e.
Spherical-harmonic expansion)；D=(θ, φ) measures position d shown in representing, wherein horizontal angle is θ, elevation angle φ.
S122：The square error of Spatial Difference minimum phase head-position difficult labor amplitude and its spheric harmonic function is carried out
L1 is regular, the model minimized the error.
It is regular by the square error progress L1 to model in this step, model error minimization problem is obtained, i.e.,：
Wherein, L is the number of each measurement point monaural HRTF amplitudes；S represents to measure the total number of position；dsRepresent s
Horizontal angle and the elevation angle at a position, ds=(θs,φs)；For the humorous spreading coefficient of sparse ball；Nm=(N0+1)2, wherein N0It is fair
Perhaps maximal degree, NmA basic function isλiFor rule
The integralization factor, the degree of rarefication of Controlling model, value range are 0~1；T(ds) represent to measure position dsLocate sound wave and reach left ear and the right side
The time difference of ear；It represents to measure position dsLocate the Spatial Difference minimum phase amplitude of left i-th of frequency band of ear；It represents to measure position dsLocate the Spatial Difference minimum phase amplitude of i-th of frequency band of auris dextra.
It is above-mentionedWithIt can be obtained by the formula used in step S113.
The embodiment of the present invention is regular by carrying out L1 to model error minimization problem, automatically removes smaller to modeling contribution
Coefficient (i.e. to rebuild head-position difficult labor and measure head-position difficult labor spectrum between the smaller system of lag effects
Number), the interference in the head-position difficult labor of measurement is reduced, the key for remaining head-position difficult labor to the full extent is special
Sign thus greatly reduces the interpolation spectrum distortion degree of the continuous HRTF of the total space.
S123：The optimal degree of rarefication of model is obtained by K cross validation method, so as to obtain sparse spherical harmonic coefficient.
Specifically, K subsample is averagely divided into all data, an individual subsample is kept as verifying
The data of model, and other K-1 sample is used for training.Then, it repeats K times, each subsample verification is primary, K times average
Result as optimal λi.Finally, all data are calculated using LASSO algorithms, so as to obtain sparse spherical harmonic coefficient,
Namely the sparse humorous spreading coefficient of ball.The sparse specific steps for solving calculating, which are carried out, using LASSO algorithms can be found in related existing skill
Art, details are not described herein.
In some embodiments, step S130 may further include：Step S131 to step S133.
S131：The continuous head correlation of minimum phase measured on position is rebuild by sparse spherical harmonic coefficient according to the following formula to pass
The time difference that defeated function amplitude estimation reaches left ear and auris dextra with sound wave is estimated：
Wherein,Represent dsLocate the left and right continuous head-position difficult labor of ear Spatial Difference minimum phase of i-th of frequency band
The estimation of amplitude；Represent l from setMiddle value, l=0 ..., Nm；Represent the N of i-th of frequency bandmIn a spherical harmonic coefficient
It is not the position where 0 coefficient；It represents to measure position dsLocate the time difference estimation that sound wave reaches left ear and auris dextra；L takes
Positive integer；Represent sparse spherical harmonic coefficient；Yl(ds) represent to measure position ds(it can pass through step S121 to the spheric harmonic function at place
In the formula used obtain).
S132：The continuous head-position difficult labor of left and right ear is determined according to the following formula：
Wherein,Represent the continuous head-position difficult labor of left ear；Represent that the continuous head of auris dextra is related
Transfer function；T0Represent that sound wave reaches the time of auris dextra, T0=Lr/ v, wherein, LrRepresent distance of the sound wave apart from auris dextra, v is represented
The velocity of sound, it is preferable that velocity of sound 340m/s.
S133：According to the different orientation in space, determine the continuous head-position difficult labor of left and right ear, obtain the company of the total space
Continuous head-position difficult labor.
This step is according to different azimuthal point ds, can finally obtain the continuous head-position difficult labor in total space orientation.
In addition, the embodiment of the present invention can also include step S140.
S140：The continuous head-position difficult labor of the total space is assessed.
This step assesses the performance of the sparse modeling of the head-position difficult labor based on spheric harmonic function.
Assessment mode includes but not limited to logarithm spectrum distortion and is assessed with amount of storage.
In practical applications, logarithmic spectrum distortion evaluation can be carried out according to the following formula：
Wherein, it measures and counts out in S representation spaces, NfRepresent number of sub-bands；Represent the continuous head phase of left and right ear
Close transfer function；H(d,fk) represent to measure obtained left and right ear head-position difficult labor；fkRepresent k-th of frequency band；k1And k2Point
The frequency range for not representing comparison is from kth1A frequency band is to kth2A frequency band；Represent all measurement head-position difficult labors
Spatial position set.
Amount of storage assessment includes absolute log spectrum distortion and opposite logarithm spectrum distortion.Wherein, absolute log spectrum distortion defines
The ratio between database that model number for reduction is stored with original needs.With respect to logarithm spectrum distortion modeling side is based on for tradition
Method obtains intuitive performance comparison.
By taking the above embodiment, the embodiment of the present invention proposes a kind of complexity for fully assessing algorithm and accurate
The assessment mode of degree.
Although each step is described in the way of above-mentioned precedence in above-described embodiment, this field
Technical staff is appreciated that the effect in order to realize the present embodiment, is performed between different steps not necessarily in such order,
It (parallel) execution simultaneously or can be performed with reverse order, these simple variations all protection scope of the present invention it
It is interior.
Based on the technical concept identical with embodiment of the method, the embodiment of the present invention also provides a kind of head based on spheric harmonic function
The sparse modeling of related transfer function.The system 20 can include：Acquisition module 22, processing module 24, modeling module 26
With generation module 28.Wherein, acquisition module 22 is used to obtain head-position difficult labor to be measured.Processing module 24 is with obtaining mould
Block 22 is connected, and for handling head-position difficult labor to be measured, generates minimum phase head-position difficult labor, and remove complete survey
The minimum phase amplitude equalizing value in direction is measured, obtains Spatial Difference minimum phase head-position difficult labor.Modeling module 26 and place
It manages module 24 to be connected, for being modeled to Spatial Difference minimum phase head-position difficult labor, obtains sparse spherical harmonic coefficient.
Generation module 28 is connected to generate the total space by sparse spherical harmonic coefficient interpolation, and according to the orientation in space with modeling module 26
Continuous head-position difficult labor.
In some embodiments, processing module 24 specifically includes：The general character component acquisition of first removal module, directional correlation
Module and the second removal module.Wherein, the first removal module is used to remove non-minimum phase in head-position difficult labor to be measured
Bit position obtains minimum phase head-position difficult labor.The general character component acquisition module of directional correlation, with the first removal module phase
Even, for calculating the minimum phase amplitude equalizing value of full measurement direction according to the following formula, the general character component of directional correlation is obtained：
Wherein, djRepresent the horizontal angle at j-th of position and the elevation angle；S represents to measure the total number of position；|Hmin(dj,fi)
| represent the amplitude of minimum phase head-position difficult labor；fiRepresent i-th of frequency band；I takes positive integer；Havg(fi) represent direction phase
The general character component of pass.Second removal module is connected with the general character component acquisition module of directional correlation, for being existed according to the following formula
In the measurement orientation of each head-position difficult labor, the general character component of directional correlation is removed from minimum phase log-magnitude, is obtained
To Spatial Difference minimum phase head-position difficult labor：
Hp(ds,fi)=20log10|Hmin(ds,fi)|-Havg(fi),
Wherein, dsRepresent the horizontal angle at s-th of position and the elevation angle；|Hmin(ds,fi) | represent that minimum phase head correlation passes
The amplitude of defeated function；Havg(fi) represent directional correlation general character component；Hp(ds,fi) represent s-th of position, i-th of frequency band fi's
Spatial Difference minimum phase amplitude.
In some embodiments, modeling module specifically includes：Computing module, regular module and degree of rarefication acquisition module.Its
In, computing module is used to calculate the spheric harmonic function at setting measurement position according to the following formula：
Wherein, l represents the number of degrees of the spheric harmonic function；M represents the exponent number of spheric harmonic function；N represents the degree of Legendre function
Number；The expression number of degrees are n, the Legendre function that exponent number is m；Represent to measure position as the humorous letter of ball at d
Number；D=(θ, φ).Regular module is connected with computing module, for Spatial Difference minimum phase head-position difficult labor width
Spend, the model that is minimized the error regular with the square error of its spheric harmonic function progress L1.Degree of rarefication acquisition module with it is regular
Module is connected, and the optimal degree of rarefication of model is obtained for passing through K cross validation method, so as to obtain sparse spherical harmonic coefficient.
In some embodiments, generation module specifically includes：Rebuild module, the first determining module and the second determining module.
Wherein, module is rebuild to be used to rebuild the continuous head phase of minimum phase measured on position by sparse spherical harmonic coefficient according to the following formula
Close the time difference estimation that transfer function amplitude Estimation reaches left ear and auris dextra with sound wave：
Wherein,Represent the measurement position dsLocate the left and right continuous head of ear Spatial Difference minimum phase of i-th of frequency band
The estimation of related transfer function amplitude；Represent l from setMiddle value, l=0 ..., Nm；L takes positive integer；Represent i-th
The N of a frequency bandmIt is not the position where 0 coefficient in a spherical harmonic coefficient；Represent the measurement position dsLocate sound wave to reach
The time difference of left ear and auris dextra is estimated；Represent sparse spherical harmonic coefficient；Yl(ds) represent the measurement position dsThe humorous letter of ball at place
Number.First determining module is connected with rebuilding module, for determining the continuous head-position difficult labor of left and right ear according to the following formula：
Wherein,Represent left ear head-position difficult labor；Represent auris dextra head-position difficult labor；
T0Represent that sound wave reaches the time of auris dextra, T0=Lr/ v, wherein, LrRepresent distance of the sound wave apart from auris dextra, v represents the velocity of sound.Second
Determining module is connected with the first determining module, for according to the different orientation in space, determining the continuous head associated transport letter of left and right ear
Number, obtains the continuous head-position difficult labor of the total space.
In some embodiments, above system can also include：Logarithmic spectrum distortion evaluation module and logarithmic spectrum distortion evaluation
Module.Wherein, logarithmic spectrum distortion evaluation module is connected with generation module, for being passed according to the following formula to the continuous head correlation of the total space
Defeated function carries out logarithmic spectrum distortion evaluation：
Wherein, it measures and counts out in S representation spaces, NfRepresent number of sub-bands；Represent the continuous head phase of left and right ear
Close transfer function；H(d,fk) represent to measure obtained left and right ear head-position difficult labor；fkRepresent k-th of frequency band；k1And k2Point
The frequency range for not representing comparison is from kth1A frequency band is to kth2A frequency band；Represent all measurement head-position difficult labors
Spatial position set.Logarithmic spectrum distortion evaluation module is connected with generation module, for the continuous head associated transport to the total space
Function carries out absolute log spectrum distortion and opposite logarithmic spectrum distortion evaluation.
It should be noted that the sparse modeling system for the head-position difficult labor based on spheric harmonic function that above-described embodiment provides
System is only carried out with the division of above-mentioned each function module for example, in practical applications when being modeled, can be as needed
And complete above-mentioned function distribution by different function modules, i.e., the module in the embodiment of the present invention or step are decomposed again
Or combination, for example, the module of above-described embodiment can be merged into a module, multiple submodules can also be further split into
Block, to complete all or part of function described above.For module, the title of step involved in the embodiment of the present invention,
Just for the sake of distinguishing modules or step, it is not intended as inappropriate limitation of the present invention.
It will be understood by those skilled in the art that the sparse modeling of the above-mentioned head-position difficult labor based on spheric harmonic function
Can also include some other known features, such as processor, controller, memory and bus etc., wherein, memory include but
It is not limited to random access memory, flash memory, read-only memory, programmable read only memory, volatile memory, non-volatile memories
Device, serial storage, parallel storage or register etc., processor include but not limited to CPLD/FPGA, DSP, arm processor,
MIPS processors etc., in order to unnecessarily obscure embodiment of the disclosure, these well known structures are not shown in FIG. 2.
It should be understood that the quantity of the modules in Fig. 2 is only schematical.According to actual needs, each module can be with
With arbitrary quantity.
Above system embodiment can be used for performing above method embodiment, technical principle, the technical issues of solving
And the technique effect generated is similar, person of ordinary skill in the field can be understood that, convenience and letter for description
Clean, the specific work process of the system of foregoing description and related explanation can refer to the corresponding process in preceding method embodiment,
Details are not described herein.
It should be pointed out that the system embodiment and embodiment of the method for the present invention are described respectively above, but it is right
The details of one embodiment description can also be applied to another embodiment.
The technical solution provided above the embodiment of the present invention is described in detail.Although it applies herein specific
A example the principle of the present invention and embodiment are expounded, still, the explanation of above-described embodiment is only applicable to help to manage
Solve the principle of the embodiment of the present invention；Meanwhile to those skilled in the art, embodiment according to the present invention, is being embodied
It can be made a change within mode and application range.
It should be noted that the flow chart or block diagram that are referred to herein are not limited solely to form shown in this article,
Other can also be carried out to divide and/or combine.
It should be noted that：Label and word in attached drawing are intended merely to be illustrated more clearly that the present invention, are not intended as pair
The improper restriction of the scope of the present invention.
Term " comprising " or any other like term are intended to cover non-exclusive inclusion, so that including a system
Process, method, article or the equipment/device of row element not only includes those elements, but also including being not explicitly listed
Other elements either further include these processes, method, article or the intrinsic element of equipment/device.
As used herein, term " module " may refer to the software object performed on a computing system or routine.
Disparate modules described herein can be embodied as to the object performed on a computing system or process (for example, as independent
Thread).While it is preferred that realize system and method described herein with software, but with hardware or software and hard
The realization of the combination of part is also possible and can be conceived to.
Term " first ", " second " etc. are for distinguishing similar object rather than for describing or representing specific suitable
Sequence or precedence.
The present invention each step can be realized with general computing device, for example, they can concentrate on it is single
On computing device, such as：Personal computer, server computer, handheld device or portable device, laptop device or more
Processor device can also be distributed on the network that multiple computing devices are formed, they can be to be different from sequence herein
They are either fabricated to each integrated circuit modules or will be more in them by the step shown or described by performing respectively
A module or step are fabricated to single integrated circuit module to realize.Therefore, the present invention is not limited to any specific hardware and soft
Part or its combination.
Method provided by the invention can also be realized using programmable logic device, can also be embodied as computer program
(it includes routines performing specific tasks or implementing specific abstract data types, program, object, component for software or program module
Or data structure etc.), such as can be according to an embodiment of the invention a kind of computer program product, run the computer journey
Sequence product makes computer perform for demonstrated method.The computer program product includes computer readable storage medium,
Comprising computer program logic or code section on the medium, it is used to implement the method.The computer readable storage medium
Can be the built-in medium being mounted in a computer or the removable medium (example that can be disassembled from basic computer
Such as：Using the storage device of hot plug technology).The built-in medium includes but not limited to rewritable nonvolatile memory,
Such as：RAM, ROM, flash memory and hard disk.The removable medium includes but not limited to：Optical storage media (such as：CD-
ROM and DVD), magnetic-optical storage medium (such as：MO), magnetic storage medium (such as：Tape or mobile hard disk), have it is built-in can
Rewrite nonvolatile memory media (such as：Storage card) and with built-in ROM media (such as：ROM boxes).
It shall also be noted that the language used in this specification primarily to readable and introduction purpose and select,
Rather than it is selected to explain or limit subject of the present invention.
Present invention is not limited to the embodiments described above, and in the case of without departing substantially from substantive content of the present invention, this field is common
Any deformation, improvement or the replacement that technical staff is contemplated that each fall within protection scope of the present invention.
Claims (10)
1. a kind of sparse modeling method of the head-position difficult labor based on spheric harmonic function, which is characterized in that the method is at least
Including：
Obtain the head-position difficult labor to be measured；
The head-position difficult labor to be measured is handled, generates minimum phase head-position difficult labor, and removes full measurement side
To minimum phase amplitude equalizing value, obtain Spatial Difference minimum phase head-position difficult labor；
The Spatial Difference minimum phase head-position difficult labor is modeled, obtains sparse spherical harmonic coefficient；
By the sparse spherical harmonic coefficient interpolation, and according to the continuous head-position difficult labor of the orientation in the space generation total space.
2. according to the method described in claim 1, it is characterized in that, the processing head-position difficult labor to be measured,
Minimum phase head-position difficult labor is generated, and removes the minimum phase amplitude equalizing value of full measurement direction, obtains Spatial Difference
Minimum phase head-position difficult labor, specifically includes：
Non-minimum phase bit position in the head-position difficult labor to be measured is removed, obtains minimum phase head associated transport letter
Number；
The minimum phase amplitude equalizing value of full measurement direction is calculated according to the following formula, obtains the general character component of directional correlation：
Wherein, the djRepresent the horizontal angle at j-th of position and the elevation angle；The S represents to measure the total number of position；It is described |
Hmin(dj,fi) | represent the amplitude of the minimum phase head-position difficult labor；The fiRepresent i-th of frequency band；The i takes just
Integer；The Havg(fi) represent the general character component of the directional correlation；
According to the following formula in the measurement orientation of each head-position difficult labor, from minimum phase log-magnitude described in removal
The general character component of directional correlation obtains Spatial Difference minimum phase head-position difficult labor：
Hp(ds,fi)=20log10|Hmin(ds,fi)|-Havg(fi),
Wherein, the dsRepresent the horizontal angle at s-th of position and the elevation angle；It is described | Hmin(ds,fi) | represent the minimum phase
The amplitude of head-position difficult labor；The Havg(fi) represent the general character component of the directional correlation；The Hp(ds,fi) represent the
S position, i-th of frequency band fiSpatial Difference minimum phase amplitude.
3. according to the method described in claim 2, it is characterized in that, described pass the Spatial Difference minimum phase head correlation
Defeated function is modeled, and is obtained sparse spherical harmonic coefficient, is specifically included：
Spheric harmonic function at setting measurement position is calculated according to the following formula：
Wherein, the l represents the number of degrees of the spheric harmonic function；The m represents the exponent number of the spheric harmonic function；The n represents to strangle
Allow the number of degrees of moral function；It is describedRepresent the Legendre function that the number of degrees are n, the exponent number is m；It is describedRepresent the measurement position for the spheric harmonic function at d；The d=(θ, φ) measures position d, wherein institute shown in representing
It states θ and represents horizontal angle, the φ represents the elevation angle；
The square error of the Spatial Difference minimum phase head-position difficult labor amplitude and its spheric harmonic function is carried out
L1 is regular, the model minimized the error；
The optimal degree of rarefication of the model is obtained by K cross validation method, so as to obtain sparse spherical harmonic coefficient.
4. according to the method described in claim 3, it is characterized in that, described by the sparse spherical harmonic coefficient interpolation, and according to
The continuous head-position difficult labor of the orientation generation total space in space, specifically includes：
The continuous head associated transport letter of minimum phase measured on position is rebuild by the sparse spherical harmonic coefficient according to the following formula
The time difference that number amplitude Estimation reaches left ear and auris dextra with sound wave is estimated：
Wherein, it is describedRepresent the measurement position dsLocate the left and right continuous head of ear Spatial Difference minimum phase of i-th of frequency band
The estimation of related transfer function amplitude；It is describedRepresent l from setMiddle value, l=0 ..., Nm；It is describedIt represents i-th
The N of frequency bandmIt is not the position where 0 coefficient in a spherical harmonic coefficient；It is describedRepresent the measurement position dsPlace's sound wave arrives
Estimate up to the time difference of left ear and auris dextra；It is describedRepresent the sparse spherical harmonic coefficient；The Yl(ds) represent the measurement position
Put dsThe spheric harmonic function at place；The L takes positive integer；
The continuous head-position difficult labor of left and right ear is determined according to the following formula：
Wherein, it is describedRepresent left ear head-position difficult labor；It is describedRepresent auris dextra head associated transport letter
Number；The T0Represent that sound wave reaches the time of auris dextra, T0=Lr/ v, wherein, the LrRepresent distance of the sound wave apart from auris dextra, institute
It states v and represents the velocity of sound；WithThe minimum phase estimation of left and right ear is represented respectively,
According to the different orientation in space, determine the continuous head-position difficult labor of left and right ear, obtain the continuous of the total space
Head-position difficult labor.
5. according to the method described in claim 1, it is characterized in that, the method further includes：
Logarithmic spectrum distortion evaluation is carried out to the continuous head-position difficult labor of the total space according to the following formula：
Wherein, it measures and counts out in the S representation spaces, the NfRepresent number of sub-bands；It is describedRepresent described left and right
The continuous head-position difficult labor of ear；H (d, the fk) represent to measure the obtained left and right ear head-position difficult labor；The fk
Represent k-th of frequency band；The k1With the k2The frequency range for representing comparison respectively is from kth1A frequency band is to kth2A frequency band；
It is describedRepresent the spatial position set of all measurement head-position difficult labors；
Absolute log spectrum distortion and opposite logarithmic spectrum distortion evaluation are carried out to the continuous head-position difficult labor of the total space.
6. a kind of sparse modeling of the head-position difficult labor based on spheric harmonic function, which is characterized in that the system is at least
Including：
Acquisition module, for obtaining the head-position difficult labor to be measured；
Processing module is connected with the acquisition module, for handling the head-position difficult labor to be measured, generates minimum phase
Potential head related transfer function, and the minimum phase amplitude equalizing value of full measurement direction is removed, obtain Spatial Difference minimum phase head
Related transfer function；
Modeling module is connected with the processing module, for the Spatial Difference minimum phase head-position difficult labor into
Row modeling, obtains sparse spherical harmonic coefficient；
Generation module is connected with the modeling module and is used for through the sparse spherical harmonic coefficient interpolation, and according to the orientation in space
Generate the continuous head-position difficult labor of the total space.
7. system according to claim 6, which is characterized in that the processing module specifically includes：
First removal module：For removing non-minimum phase bit position in the head-position difficult labor to be measured, minimum is obtained
Phase head-position difficult labor；
The general character component acquisition module of directional correlation is connected with the described first removal module, complete for being calculated according to the following formula
The minimum phase amplitude equalizing value of measurement direction obtains the general character component of directional correlation：
Wherein, the djRepresent the horizontal angle at j-th of position and the elevation angle；The S represents to measure the total number of position；It is described |
Hmin(dj,fi) | represent the amplitude of the minimum phase head-position difficult labor；The fiRepresent i-th of frequency band；The i takes just
Integer；The Havg(fi) represent the general character component of the directional correlation；
Second removal module, be connected with the general character component acquisition module of the directional correlation, for according to the following formula each
In the measurement orientation of head-position difficult labor, the general character component of the directional correlation is removed from minimum phase log-magnitude, is obtained
To Spatial Difference minimum phase head-position difficult labor：
Hp(ds,fi)=20log10|Hmin(ds,fi)|-Havg(fi),
Wherein, the dsRepresent the horizontal angle at s-th of position and the elevation angle；It is described | Hmin(ds,fi) | represent the minimum phase
The amplitude of head-position difficult labor；The Havg(fi) represent the general character component of the directional correlation；The Hp(ds,fi) represent the
S position, i-th of frequency band fiSpatial Difference minimum phase amplitude.
8. system according to claim 7, which is characterized in that the modeling module specifically includes：
Computing module, for calculating the spheric harmonic function at setting measurement position according to the following formula：
Wherein, the l represents the number of degrees of the spheric harmonic function；The m represents the exponent number of the spheric harmonic function；The n represents to strangle
Allow the number of degrees of moral function；It is describedRepresent the Legendre function that the number of degrees are n, the exponent number is m；It is describedRepresent the measurement position for the spheric harmonic function at d；The d=(θ, φ) measures position d shown in representing, wherein described
θ represents horizontal angle, and the φ represents the elevation angle；
Regular module is connected with the computing module, for the Spatial Difference minimum phase head-position difficult labor width
Spend, the model that is minimized the error regular with the square error progress L1 of spheric harmonic function its described；
Degree of rarefication acquisition module is connected with the regular module, optimal for passing through the K cross validation method acquisition model
Degree of rarefication, so as to obtain sparse spherical harmonic coefficient.
9. system according to claim 8, which is characterized in that the generation module specifically includes：
Module is rebuild, it is continuous for rebuilding the minimum phase measured on position by the sparse spherical harmonic coefficient according to the following formula
The time difference that head-position difficult labor amplitude Estimation reaches left ear and auris dextra with sound wave is estimated：
Wherein, it is describedRepresent the measurement position dsLocate the left and right continuous head of ear Spatial Difference minimum phase of i-th of frequency band
The estimation of related transfer function amplitude；It is describedRepresent l from setMiddle value, l=0 ..., Nm；It is describedIt represents i-th
The N of frequency bandmIt is not the position where 0 coefficient in a spherical harmonic coefficient；It is describedRepresent the measurement position dsPlace's sound wave arrives
Estimate up to the time difference of left ear and auris dextra；It is describedRepresent the sparse spherical harmonic coefficient；The Yl(ds) represent the measurement position
Put dsThe spheric harmonic function at place；The L takes positive integer；
First determining module is connected with the reconstruction module, for determining the continuous head associated transport of left and right ear according to the following formula
Function：
Wherein, it is describedRepresent left ear head-position difficult labor；It is describedRepresent auris dextra head associated transport letter
Number；The T0Represent that sound wave reaches the time of auris dextra, T0=Lr/ v, wherein, the LrRepresent distance of the sound wave apart from auris dextra, institute
It states v and represents the velocity of sound；WithThe minimum phase estimation of left and right ear is represented respectively,
Second determining module is connected with first determining module, for according to the different orientation in space, determining described left and right
The continuous head-position difficult labor of ear obtains the continuous head-position difficult labor of the total space.
10. system according to claim 6, which is characterized in that the system also includes：
Logarithmic spectrum distortion evaluation module is connected with the generation module, for according to the following formula to the continuous head phase of the total space
It closes transfer function and carries out logarithmic spectrum distortion evaluation：
Wherein, it measures and counts out in the S representation spaces, the NfRepresent number of sub-bands；It is describedRepresent described left and right
The continuous head-position difficult labor of ear；H (d, the fk) represent to measure the obtained left and right ear head-position difficult labor；The fk
Represent k-th of frequency band；The k1With the k2The frequency range for representing comparison respectively is from kth1A frequency band is to kth2A frequency band；
It is describedRepresent the spatial position set of all measurement head-position difficult labors；
Logarithmic spectrum distortion evaluation module is connected with the generation module, for the continuous head associated transport letter to the total space
Number carries out absolute log spectrum distortion and opposite logarithmic spectrum distortion evaluation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610802607.0A CN106412793B (en) | 2016-09-05 | 2016-09-05 | The sparse modeling method and system of head-position difficult labor based on spheric harmonic function |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610802607.0A CN106412793B (en) | 2016-09-05 | 2016-09-05 | The sparse modeling method and system of head-position difficult labor based on spheric harmonic function |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106412793A CN106412793A (en) | 2017-02-15 |
| CN106412793B true CN106412793B (en) | 2018-06-12 |
Family
ID=57998376
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610802607.0A Active CN106412793B (en) | 2016-09-05 | 2016-09-05 | The sparse modeling method and system of head-position difficult labor based on spheric harmonic function |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106412793B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107301153B (en) * | 2017-05-22 | 2021-05-07 | 上海大学 | Head-related transfer function modeling method based on self-adaptive Fourier decomposition |
| CN107480100B (en) * | 2017-07-04 | 2020-02-28 | 中国科学院自动化研究所 | Head-related transfer function modeling system based on deep neural network intermediate layer characteristics |
| CN107942314B (en) * | 2017-11-22 | 2021-06-04 | 中南大学 | Doppler through-wall radar positioning method based on LASSO feature extraction |
| CN108596016B (en) * | 2018-03-06 | 2021-11-09 | 北京大学 | Personalized head-related transfer function modeling method based on deep neural network |
| CN109145360B (en) * | 2018-06-29 | 2020-08-04 | 中国科学院自动化研究所 | Head-related transfer function personalized modeling system based on sparse constraint |
| CN110412333B (en) * | 2019-04-30 | 2020-10-13 | 清华大学 | Current parameter elastic network regularization inversion method based on spherical harmonic decomposition |
| CN110489470B (en) * | 2019-07-16 | 2022-11-29 | 西北工业大学 | HRTF (head related transfer function) personalization method based on sparse representation classification |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2899423A1 (en) * | 2006-03-28 | 2007-10-05 | France Telecom | Three-dimensional audio scene binauralization/transauralization method for e.g. audio headset, involves filtering sub band signal by applying gain and delay on signal to generate equalized and delayed component from each of encoded channels |
| US9420393B2 (en) * | 2013-05-29 | 2016-08-16 | Qualcomm Incorporated | Binaural rendering of spherical harmonic coefficients |
| US9749769B2 (en) * | 2014-07-30 | 2017-08-29 | Sony Corporation | Method, device and system |
-
2016
- 2016-09-05 CN CN201610802607.0A patent/CN106412793B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN106412793A (en) | 2017-02-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106412793B (en) | The sparse modeling method and system of head-position difficult labor based on spheric harmonic function | |
| Lehmann et al. | Diffuse reverberation model for efficient image-source simulation of room impulse responses | |
| Breckling | The analysis of directional time series: applications to wind speed and direction | |
| CN105083320B (en) | The detection method and device of rail smooth state | |
| CN107480100B (en) | Head-related transfer function modeling system based on deep neural network intermediate layer characteristics | |
| van Lieshout | On estimation of the intensity function of a point process | |
| Váša et al. | Mesh statistics for robust curvature estimation | |
| CN103699678A (en) | Hierarchical clustering method and system based on multistage layered sampling | |
| WO2023116196A1 (en) | Aoa and tof joint estimation method and apparatus for indoor positioning, and storage medium | |
| Soderstrom | Accuracy analysis of the Frisch scheme for identifying errors-in-variables systems | |
| Tang et al. | Learning acoustic scattering fields for dynamic interactive sound propagation | |
| CN112885375A (en) | Global signal-to-noise ratio estimation method based on auditory filter bank and convolutional neural network | |
| CN104503963B (en) | Head coherent pulse response data sets processing method | |
| Gaultier et al. | VAST: The virtual acoustic space traveler dataset | |
| Barumerli et al. | Round Robin Comparison of Inter-Laboratory HRTF Measurements–Assessment with an auditory model for elevation | |
| Ick et al. | Blind acoustic room parameter estimation using phase features | |
| Kaneko et al. | Ear shape modeling for 3D audio and acoustic virtual reality: The shape-based average HRTF | |
| She et al. | A new adaptive local linear prediction method and its application in hydrological time series | |
| CN104318064B (en) | Head coherent pulse response three-dimensional data compression method based on model's Multidimensional decomposition technique | |
| Yoshimoto et al. | Constructing tree stem form from digitized surface measurements by a programming approach within discrete mathematics | |
| Georgiou et al. | Robust maximum likelihood source localization: The case for sub-Gaussian versus Gaussian | |
| US8942966B2 (en) | Method for parameterizing and morphing stochastic reservoir models | |
| Warnecke et al. | HRTF personalization based on ear morphology | |
| CN107423513A (en) | A kind of method, equipment and its storage device of four-dimensional interpolation data processing | |
| Song et al. | Central limit theorems of local polynomial threshold estimator for diffusion processes with jumps |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |