CN102984642A - Three-dimensional translation method for five loudspeakers - Google Patents

Abstract

The invention provides a three-dimensional translation method for five loudspeakers. The method comprises the steps of judging a position relation among a virtual sound source (substituted single loudspeaker) and five substituting loudspeakers, measuring and recording positions of all the loudspeakers, calculating a distribution coefficient of a signal of the substituted single loudspeaker being distributed to signals of the five substituting loudspeakers according to a principle that time average sound intensity and sound pressure generated by the virtual sound source (the substituted single loudspeaker) at a receiving point are equal to average sound intensity and sound pressure generated by the five substituting loudspeakers at the receiving point, deleting the virtual sound source, and distributing the signals. According to the three-dimensional translation method for the five loudspeakers, the five substituting loudspeakers can substitute one virtual sound source (the substituted loudspeaker), and compared with the translation method for fewer loudspeakers, repeated adjustments of grouping and signal distribution under the condition of the five loudspeakers are avoided, and the signal distribution can be conducted rapidly under the condition of the five loudspeakers.

Description

A kind of D translation method of five loud speakers

Technical field

The invention belongs to field of acoustics, relate in particular to a kind of D translation method of five loud speakers.

Background technology

With respect to 5.1 audio systems, some multi-channel sound systems strengthen the spatial sound picture, and this some of them system can provide three dimensional sound.Yet, only have the development of attempting few in number to be used for the three-dimensional audio content among these systems.A difficulty that produces three-dimensional audio is when using traditional sound reconstructing system, the time that the D translation cost is a large amount of.D translation is a kind of technology of using some loud speakers manufacturing virtual sound sources in three-dimensional sound field.In the D translation method, extensively approved based on the amplitude translation technology (vector based amplitude panning brief note is VBAP) of vector.In three-dimensional VBAP, virtual sound source is synthetic with three loud speakers, and starting point is at acceptance point, and terminal point can represent that (each vectorial starting point is at acceptance point with other three SYSTEM OF LINEAR VECTOR at the unit vector of virtual sound source position, terminal point is in each vector institute corresponding loud speaker position, and length is 1).The coefficient of three vector representations represents vector corresponding three loud speakers as weight coefficient so that the signal of virtual sound source is assigned to through after the normalization.If the loud speaker number that the expression virtual sound source uses is more than three, the VBAP technology is several subspaces with whole reconstruction speaker volume by three loud speaker one components, carries out signal according to VBAP and distribute in every sub spaces.The AkioAndo in NHK laboratory in 2007 has proposed another D translation method, its basic thought is in multi-channel system, realizes that single loud speaker (regarding virtual sound source as) is constant in the sound physical property at acceptance point place with three loud speakers that replace virtual sound source in the sound physical property at acceptance point place.He is self-defined acoustic pressure vector is as physical property.Condition is replaced loud speaker (virtual sound source) and the direction of the vector (starting point is at acceptance point, and terminal point is in replaced loud speaker position) of acceptance point formation consist of spherical triangles three alternative set of speakers inside (center that sphere is corresponding is acceptance point).When distance between all loud speakers and the acceptance point equated, the result that Akio Ando has proposed the D translation method was the same with the result of VBAP, and can be considered is the popularization of VBAP technology.The Akio Ando in NHK laboratory in 2009 has proposed again a kind of D translation method, and its basic thought is the same with the D translation method that he proposed in 2007, but the physical property of taking is time average intensity of sound and acoustic pressure.The time average intensity of sound refers to that voice signal is by the average acoustical power of unit are in a period of time, and it is a vector, can be good at representing sound localization.Akio Ando has proved that also the result who utilizes the alternative virtual sound source (single loud speaker) of three set of speakers to obtain by this kind method is the same with the D translation method of VBAP method and proposition in 2007, but VBAP method, it is that three alternative set of speakers substitute virtual sound sources (single loud speaker) that the D translation method of the D translation method of Akio Ando proposition in 2007 and Akio Ando proposition in 2009 all needs, when the loud speaker number is more, must carry out the loud speaker grouping, be that per three loud speakers are divided into one group and substitute virtual sound source, trouble comparatively.When alternative set of speakers number is four, for fear of per three loud speakers grouping, Akio Ando has released the situation of using four set of speakers to substitute a virtual sound source (single loud speaker) by the D translation method that proposes, so that the signal distribution becomes simple and easy to do in actual applications.But the practical application scene is extremely complicated, when alternative loud speaker number be five and more than when being, the method of four alternative virtual sound sources (single loud speaker) of loud speaker is easy not again, so we need to seek the more system for delivering of the alternative virtual sound source (single loud speaker) of multi-loudspeaker.

Summary of the invention

The present invention is directed to the deficiencies in the prior art, a kind of D translation method of five loud speakers is provided.

Technical solution of the present invention provides a kind of D translation method of five loud speakers, five loud speakers replacing in the set of speakers all are placed on the same sphere, the single loud speaker that is replaced also is positioned on the sphere of replacing the set of speakers place, the single loud speaker that is replaced is considered as virtual sound source, and the sphere center position of sphere is acceptance point; Realize that D translation comprises following steps,

Step 1 is judged virtual sound source and the position relationship of replacing set of speakers, confirms that virtual sound source is positioned at five replacement loud speakers and consists of the spheric polygon intra-zone;

Step 2 is carried out loudspeaker position and is measured and record, comprises take acceptance point as origin of coordinates O sets up three-dimensional cartesian coordinate system XYZ, measures the polar coordinates of single loud speaker Sp0 in three-dimensional cartesian coordinate system XYZ that are replaced and is designated as

All the other five replacement loud speaker Sp1, Sp2, Sp3, Sp4, the Sp5 polar coordinates in three-dimensional cartesian coordinate system XYZ are designated as respectively

Step 3, the distribution coefficient w of five replacement loud speakers of calculating ₁, w ₂, w ₃, w ₄, w ₅, establish w ₄=g ₁, w ₅=g ₂, calculate according to following formula and realize, $w_1=\frac{D_1}{D},$ $w_2=\frac{D_2}{D},$ $w_3=\frac{D_3}{D},$ w ₄＝g ₁,w ₅＝g ₂ (6)

D ₄=g ₁D (10)

D ₅=g ₂D (11)

Parameter g wherein ₁, g ₂Be respectively the default distribution factor of replacing loud speaker Sp4, Sp5;

Step 4, the deletion virtual sound source, with the signal times of virtual sound source with the distribution coefficient w in the step 3 ₁, w ₂, w ₃, w ₄, w ₅Be assigned in five replacement loud speakers.

Adopt the D translation method of five loud speakers provided by the invention, can replace a virtual sound source (being replaced loud speaker) with five replacement loud speakers, repeatedly adjustment compared to the shift method of three and four loud speakers has avoided grouping under five loud speaker situations and signal to distribute is conducive to carry out fast signal and distributes in the situation of five loud speakers.This patent not only can be used for the situation that a virtual sound source (being replaced loud speaker) is replaced by five replacement loud speakers, also can be used for comprising in the system of multi-loudspeaker more the situation that the method that several loud speakers are wherein used according to this patent is one by one replaced.

Description of drawings

Fig. 1 is the flow chart of the embodiment of the invention

Fig. 2 is loud speaker and the virtual sound source relative position figure of the embodiment of the invention.

Fig. 3 is that the selected replacement set of speakers of the embodiment of the invention consists of the spheric polygon schematic diagram.

Embodiment

The present invention is directed to five set of speakers that are placed on the same sphere, proposed a kind of D translation method of five loud speakers, describe technical solution of the present invention in detail below in conjunction with the drawings and specific embodiments.

As shown in Figure 1, embodiment adopts the D translation method of above-mentioned five loud speakers to realize five loud speakers replacement virtual sound sources (being replaced single loud speaker), comprises following steps:

Step 1 is judged virtual sound source (the single loud speaker that is replaced) and the position relationship of replacing set of speakers.The number of replacing loud speaker is five, virtual sound source (the single loud speaker that is replaced) need to be positioned at five replacement loud speakers and consist of the spheric polygon intra-zone, otherwise can not replace, and the area in five replacement loud speaker formation spheric polygon zones is the smaller the better.Referring to Fig. 2, hollow dots represents the position of virtual sound source (being replaced loud speaker), solid dot represents to replace the loud speaker position: replacement loud speaker Sp1, Sp2, Sp3, Sp4, Sp5 lay respectively at some C, D, E, F, the G place on ball O surface, virtual sound source (the single loud speaker that is replaced) Sp0 is positioned at the some B place on ball O surface, and some B is positioned at spheric polygon CDEFG intra-zone.Replace spheric polygon that loud speaker loca place consists of and should be the spheric polygon of replacing less in two spheric polygons that loud speaker loca place consists of, as shown in Figure 3, wherein hollow dots represents the position of virtual sound source (being replaced loud speaker), and solid dot represents to replace the loud speaker position.

Step 2, loudspeaker position is measured and record.In the replacement set of speakers in step 1, measure the position be replaced each loud speaker place in single loud speaker and five the replacement set of speakers and record;

If set up three-dimensional cartesian coordinate system XYZ take acceptance point as origin of coordinates O, the present invention adopts polar form, such as an A (ρ _A, θ _A,

In the coordinate, ρ _ADistance between expression point A and the origin of coordinates O, θ _ALine is at the projection of XOY plane and the angle between the X-axis between expression point A and the initial point O,

Angle between expression point A and the initial point O between line and the XOY plane.Suppose to measure and be replaced single loud speaker Sp0 coordinate and be

The coordinate of all the other five replacement loud speaker Sp1, Sp2, Sp3, Sp4, Sp5 is respectively

Being replaced single loud speaker Sp0 coordinate in the present embodiment is B (2,0 °, 48 °), the coordinate of all the other five replacement loud speaker Sp1, Sp2, Sp3, Sp4, Sp5 is respectively C (2,0 °, 90 °), D (2,270 °, 30 °), E (2,90 °, 30 °), F (2,315 °, 0 °), G (2,45 °, 0 °).

Step 3, distribution coefficient calculates.The signal that the principle that the time average intensity of sound that the time average intensity of sound that produces at the acceptance point place according to virtual sound source (being replaced single loud speaker) and acoustic pressure and five replacement loud speakers produce at the acceptance point place and acoustic pressure equate, calculating are replaced single loud speaker is assigned to the distribution coefficients of five replacement loudspeaker signals.

Being replaced single loud speaker in the time average intensity of sound of acceptance point place generation is:

Wherein:

ρ represents to be replaced the distance between loud speaker Sp0 position and the origin of coordinates O;

θ represents to be replaced between loud speaker Sp0 position and the initial point O line at the projection of XOY plane and the angle between the X-axis;

Expression is replaced the angle between the line and XOY plane between loud speaker Sp0 position and the initial point O;

G is illustrated in the proportionality coefficient that produces acoustic pressure with the acoustic pressure of this loud speaker of loud speaker unit distance place and speaker;

The Fourier transform of s (ω) expression loud speaker input signal.

Five time average intensities of sound that the replacement loud speaker produces at the acceptance point place

For:

θ _lExpression is replaced between loud speaker l position and the initial point O line at the projection of XOY plane and the angle between the X-axis, (l=1,2,3,4,5);

Angle between expression replacement loud speaker l position and the initial point O between line and the XOY plane, l=1,2,3,4,5;

w _l, w _jThe signal distribution coefficient of loud speaker l, j, j=1,2,3,4,5 are replaced in expression respectively.

The acoustic pressure P that the single loud speaker Sp0 that is replaced produces at the acceptance point place _BFor:

$P_B=G\frac{e^{-\mathrm{ik\ρ}}}{\mathrm{\ρ}}s\left(\mathrm{\ω}\right)---\left(3\right)$

Wherein, i is imaginary part unit, and k is wave number, F is the frequency of sound, and c is the speed that sound is propagated in air;

Five replacement loud speaker Sp1, Sp2, Sp3, Sp4, Sp5 at the acoustic pressure P ' of acceptance point place generation are:

$P^{\′}=G\underset{j=1}{\overset{5}{\mathrm{\Σ}}}\frac{e^{-\mathrm{ik\ρ}}}{\mathrm{\ρ}}{w}_{j}s\left(\mathrm{\ω}\right)---\left(4\right)$

Equated to calculate with (4) formula by (3):

w ₁+w ₂+w ₃+w ₄+w ₅=1 (5)

(1) equates with (2), and establish w ₄=g ₁, w ₅=g ₂, simultaneous (5) formula obtains: $w_1=\frac{D_1}{D},$ $w_2=\frac{D_2}{D},$ $w_3=\frac{D_3}{D},$ w ₄=g ₁,w ₅=g ₂ (6)

D ₄=g ₁D (10)D ₅=g ₂D (11)

Parameter g wherein ₁, g ₂Be the 4th (Sp4, F place), the distribution coefficient of the individual replacement loud speaker in the 5th (Sp5, G place) can artificially be set adjustment, g ₁, g ₂Excursion be 0 ~ 1, g ₁, g ₂Variation will cause w ₁, w ₂, w ₃Variation.

Suppose g in the present embodiment ₂=g ₁+ 0.1, because g ₁+ g ₂=w ₄+ w ₅≤ 1, g ₁Excursion be 0 ~ 0.4, g ₂Excursion be 0.1 ~ 0.5, g ₁, g ₂Change step all is 0.1, certainly in actual applications also can be in 0 ~ 1 scope Set arbitrarily, present embodiment is for the purpose of simplifying the description convenience just.According to formula (6) (7) (8) (9) (10) (11) (12), can calculate w ₁, w _2,w ₃, w ₄, w ₅A series of values, referring to table 1.Since the signal distribution coefficient generally get on the occasion of, get w in the table 1 so be assigned to the signal distribution coefficient of loud speaker in the present embodiment ₁, w ₂, w ₃, w ₄, w ₅Complete is a group of positive number, i.e. w ₁=0.2853, w ₂=0.1482, w ₃=0.0665, w ₄=0.2, w ₅=0.3.Therefore be assigned to C, D, E, F, the distribution coefficient of G place loudspeaker signal corresponds to respectively 0.2853,0.1482,0.0665,0.2,0.3.

Table 1 signal distribution coefficient

w 1	w 2	w 3	w 4＝g 1	w 5＝g 2
					-0.7429	0.8623	0.7806	0	0.1
-0.2288	0.5052	0.4236	0.1	0.2
					0.2853	0.1482	0.0665	0.2	0.3
0.7995	-0.2089	-0.2905	0.3	0.4
					1.3136	-0.566	-0.6476	0.4	0.5

Step 4, signal distributes, deletion virtual sound source (the single loud speaker that is replaced).With the signal times of virtual sound source (being replaced loud speaker) Sp0 with the distribution coefficient w in the step 3 ₁, w ₂, w ₃, w ₄, w ₅Be assigned among five replacement loud speaker Sp1, Sp2, Sp3, Sp4, the Sp5.

Specific embodiment described herein only is to the explanation for example of the present invention's spirit.Those skilled in the art can make various modifications or replenish or adopt similar mode to substitute described specific embodiment, but can't depart from spirit of the present invention or surmount the defined scope of appended claims.

Claims (1)

1. the D translation method of five loud speakers, it is characterized in that: five loud speakers replacing in the set of speakers all are placed on the same sphere, the single loud speaker that is replaced also is positioned on the sphere of replacing the set of speakers place, the single loud speaker that is replaced is considered as virtual sound source, and the sphere center position of sphere is acceptance point; Realize that D translation comprises following steps,

Step 1 is judged virtual sound source and the position relationship of replacing set of speakers, confirms that virtual sound source is positioned at five replacement loud speakers and consists of the spheric polygon intra-zone;

Step 2 is carried out loudspeaker position and is measured and record, comprises take acceptance point as origin of coordinates O sets up three-dimensional cartesian coordinate system XYZ, measures the polar coordinates of single loud speaker Sp0 in three-dimensional cartesian coordinate system XYZ that are replaced and is designated as

All the other five replacement loud speaker Sp1, Sp2, Sp3, Sp4, the Sp5 polar coordinates in three-dimensional cartesian coordinate system XYZ are designated as respectively

Step 3, the distribution coefficient w of five replacement loud speakers of calculating ₁, w ₂, w ₃, w ₄, w ₅, establish w ₄=g ₁, w ₅=g ₂, calculate according to following formula and realize, $w_1=\frac{D_1}{D},$ $w_2=\frac{D_2}{D},$ $w_3=\frac{D_3}{D},$ w ₄＝g ₁,w ₅=g ₂ (6)

D ₄=g ₁D (10)

D ₅=g ₂D (11)

Parameter g wherein ₁, g ₂Be respectively the default distribution factor of replacing loud speaker Sp4, Sp5;

Step 4, the deletion virtual sound source, with the signal times of virtual sound source with the distribution coefficient w in the step 3 ₁, w ₂, w ₃, w ₄, w ₅Be assigned in five replacement loud speakers.

Images