EP3061271A1

EP3061271A1 - Wave field synthesis system

Info

Publication number: EP3061271A1
Application number: EP14781270.5A
Authority: EP
Inventors: Frank Stefan SCHMIDT; Helmut Oellers
Original assignee: Advanced Acoustic SF GmbH
Current assignee: Holoplot GmbH
Priority date: 2013-08-10
Filing date: 2014-09-12
Publication date: 2016-08-31
Anticipated expiration: 2034-09-12
Also published as: WO2015036845A1; PL3061271T3; ES2674771T3; EP3061271B1; DE102013013377A1; DE112014003702A5; TR201808776T4; US9716961B2; US20160192103A1

Abstract

The present invention describes an apparatus comprising sound transducers based on the principle of wave field synthesis that can be compiled in a local design to form a freely scalable overall magnitude. The local design of the system ensures the free scalability. The computation complexity in the central system remains independent of the number of sound transducers in the overall system. In previously known apparatuses for reconstructing a sound field based on the principle of wave field synthesis, the demands on the central system rise with the number of sound transducers that it covers. For this reason too, the wave field synthesis is usually reduced to the horizontal plane of the listener, and the third dimension is lost in the reproduction of the sound events. The solution according to the invention allows the model-based approach of wave field synthesis to be also used to construct two-dimensional apparatuses of arbitrary magnitude from sound transducers based on the principle of wave field synthesis that are able to implement rapid changes of location for the virtual sound sources with the associated Doppler effects without artefacts. In this case, the requisite computation power can therefore be distributed locally because the volume of data to be transferred between the subsystems remains comparatively small.

Description

Wave field synthesis system

The present invention relates to a device of sound transducers according to the principle of wave field synthesis.

Background of the invention

With the method of wave field synthesis for the reproduction of audio signals [1], first described by Prof. Berkhout in 1988, the wave fronts emanating from a natural sound source can be physically reconstructed according to Huygens principle. From the elementary waves of a large number of individually controlled sound transducers, a virtual sound source is created at the position of the natural sound source.

When these transducers are constructed on a two-dimensional surface, the principle of wave field synthesis realizes the "acoustic curtain." In its range, all the sound sources and also the reflections of these sound sources in the recording room can be physically reconstructed in all three spatial dimensions, thus reproducing in a source-free reproduction area the acoustics of the recording room arise.

In order to completely reconstruct the acoustic conditions in the recording room, it would be necessary to build the acoustic curtain around the listener, so that all reflections of the recording room can be generated at their correct starting point. In practice, such an "acoustic cabin" has not been realized until today, since the number of sound transducers would be very large because they would have to be arranged at a small distance from one another because of the otherwise occurring aliasing effects.

In practice, the method of wave field synthesis is therefore usually reduced to a horizontal series of transducers placed around the listener. This also reduces the playback to this horizontal level, correct spatial reproduction is no longer possible. In addition, the cylindrical propagation of the wavefronts then requires that the acoustics of the playback room must be completely suppressed.

In recent years, however, some research institutions have succeeded in building a two-dimensional acoustic curtain. In the event that not all

Reflections of the recording room are reconstructed at their correct starting point, but only the psychoacoustically significant direct wavefronts and the first high-sound reflections, is described in [3] a solution that in a

C0NFIRMATI0N COPY model-based approach to replace the arranged around the listener sound transducer by deliberately generated reflections of the playback room replaced.

However, such an acoustic curtain, which is set up frontally in front of the listener, is hardly practicable if it is constructed as a whole unit. It must be big enough to reproduce the direct wavefronts in its area. The effort for it is enormous, apart from that he would hardly even as a ready assembled unit in the

Transport playback room.

For a limited number of fixedly positioned virtual sound sources, the computational effort for wave field synthesis also remains manageable for the construction of a two-dimensional acoustic curtain when several systems are coupled. However, if a sound source moves in the recording room, all run times and all levels of all dependent reflections to each individual sound transducer must be recalculated. To carry out this operation so quickly for all the acoustic curtain transducers that the movement of a sound source can be continuously displayed encounters even the model-based approach of wave field synthesis and limited to the reflections

1st order today to the limit of technical possibilities.

Even far higher is the computational effort, if a change of location of the listener in the recording room should be displayed. Then all run times of all direct wavefronts and all reflections to each individual transducer change. The newly calculated data would have to be read in by at least eight times a second

to portray reasonably fluid movement [4].

Because of the available computational power, the data - based approach uses

Wave field synthesis therefore resorted to the impulse responses for each

Predicting and storing transducers for discrete source positions in order to shift the virtual sound sources abruptly from one position to the next [5].

However, in sudden movements of the virtual sound sources are not the

Doppler effects generated by the change of location of a real sound source.

The object of the invention is therefore to describe a device that is portable for practical reasons and in the computing power in the

Central unit does not grow with the number of transducers.

So it should also be possible to change the positions of the virtual sound sources so fast that rapid movements of a sound source in three-dimensional space cause the natural Doppler effects of the change in location of a real sound source. Description of the invention

The above objects and other objects to be taken from the description are achieved by a device according to the features of claim 1. Further advantageous embodiments of the invention are specified in the dependent claims. A preferred embodiment of the present invention is in

The following drawings and detailed description, but is not intended to limit the present invention thereto.

According to the invention, the device of sound transducers according to the principle of

Wave field synthesis not as a closed unit, as described for example in [6] to execute, but to build decentralized. The individual modules are usually executed the same way. An enclosing housing may allow a modular design. This has the advantage that the modules are interchangeable and that they only have to be assigned to a position in the coordinate system during the setup process of the system. In addition, they can be preassembled and pre-wired for live PA in groups to ensure a quick setup of the system.

All audio signals can then be routed in a common line to each module. With a decentralized structure of the system, the data for the delay times and levels for each individual transducer can be transmitted very effectively with a serial transmission protocol if the model - based approach of the

Wave field synthesis is used. All audio channels of the system are then routed to all modules in a data stream.

The additional amount of data to be transmitted to the modules for the calculation of the signals for each individual transducer in a second data stream is comparatively very small. The synthesis of content, ie the audio signals themselves, and form, ie the associated data, is then no longer carried out according to the invention in a central unit, but autonomously in each modular unit. Because of the modular design, no more differentiated data or individual audio signals for each individual transducer must be transmitted. The data stream, which is fed from the central unit to all modules, only contains the vector of each virtual sound source to be displayed to a single reference point in the system.

In the modules themselves, after the setup process, the vector of a, in all modules, the same reference point of the relevant module to this common reference point is known, because it results from the edge lengths of the modules or modules and their position within the array of transducers. In the module or module itself then the vectors of each individual transducer are stored to this reference point. The vector addition of the reference point of the arrangement of sound transducers for Coordinate origin plus vector of the reference point of the module to the reference point of the arrangement of sound transducers plus vector of the respective sound transducer to the reference point of the module gives the exact position of the respective sound transducer to

Coordinate origin of the system.

In the model-based approach of wave field synthesis, the vector of each virtual sound source to the origin of coordinates is also known. Therefore, within each module, the distance between each virtual sound source to each transducer can be calculated. From this it is easy to derive the duration of the sound from this virtual sound source to the relevant sound transducer, if the correct, depending on the temperature of the propagation medium before the acoustic curtain, current

Sound velocity is known.

The number of virtual sound sources for the representation of the sources themselves and their first high-sound reflections is manageable in the model-based approach of wave field synthesis. Considering the directional resolution of human hearing, it is generally considered that there is no perceptible benefit to having more than 32 separate positions of direct sound sources [7].

If only the first high-sound reflections of these sources are synthesized correctly, seven virtual starting points of virtual sound sources are created for each audio channel. In the digital system, it is obvious to reserve an eighth position for additional high-sound reflection.

Thus, only 32 × 8 = 256 source positions for direct wavefronts and first high-sound reflections as a vectorial variable to the reference point of the arrangement of sound transducers are formed in the common data stream which is led to the assemblies. By comparison, in an acoustic curtain constructed entirely in one unit, for example with 1024 individual transducers, 32 x 8 x 1024 would be starting points of virtual

Defining sound sources in all three spatial positions is 262144 vectorial quantities. The data volume would be even higher in the data-based approach. Here not only the positions of the source and the first high-sound reflection are contained in the impulse responses, but all the starting points of all reflections are restored to the impulse response in the convolution. It would only be possible to transfer this amount of data for many individual positions of transducers on a line if there was a lot of time available for transmission. But then rapid changes in location of the sound source would not be possible.

Therefore, in the data-based method of wave field synthesis, the audio signals are convoluted into the respective impulse responses in a central processing unit, and the output of this convolution is sent to the individual output amplifiers. The larger the number of individual transducers becomes, the larger becomes the one in the

Central processing unit to be processed. This has in the past meant that the principle of wave field synthesis has almost always been reduced to a single horizontal series of transducers.

Fundamental advantage of the inventive modular structure of the arrangement of sound transducers according to the principle of wave field synthesis is that the centrally processed computing power and to be transferred to the modules amount of data is independent of the number of transducers in the overall system. It is not higher for an arbitrarily large acoustic curtain than for a single sound transducer.

This makes the representation of very fast movements of the sound source together with the associated changes in position of their first high - sound reflections in the

model-based approach possible. The eight times three distance values from the central calculation, even with an extremely high resolution of the individual values with 24 bits, only yield a data volume of 576 bits for each sound source and its first high-sound strength

Reflections. For 32 audio channels, only a data rate of 2.304 kbytes results for all position data. This value is so small that even the [6] required 8 updates per second for a smooth movement of the source can easily be easily exceeded.

Previously, because of the limited updatability of the source positions, even with the embodiments of wave-field synthesis systems reduced to the horizontal plane, it was customary to define impulse responses for predefined source positions and store them in the system to then jump the virtual sound source from one predefined position to another. As long as the source only jumps from one point to the other, it is stationary at the respective point and arises when the position changes abruptly

inevitably not the natural Doppler effects of a continuous movement that evaluates our ear very sensitively, but artifacts.

However, if the amount of data to be processed in each case in the model-based approach of wavefield synthesis is low because of the division into individual modules and the

Source positions can follow very quickly the natural location changes of the original sound sources and their first high-sound reflections, so is the representation of

Doppler effects on moving virtual sound sources without artifacts possible, because a continuous movement of the virtual sound sources in space arises.

The modular structure of the wave field synthesis system creates another fundamental advantage. Since the amount of data to be transferred and the computational effort in the central unit is independent of the number of modules or modules connected, the system becomes freely scalable. So not only the usual reduction of the Be easily overcome on the horizontal level of the listener. Even very large acoustic curtains with directional effects down to the bass range and tightly focused concave wavefronts can be realized.

Even the "acoustic cabin", which comes close to the theoretical approach of wave field synthesis described in the Kirchhoff Helmholtz Integral to a complete physical reconstruction of the output sound field in a source-free volume, would be feasible due to the free scalability of the modular system.

For this purpose, it would also be possible to construct subsystems of different performance in the modular structure, for example, to choose a larger distance between the sound transducers behind the listener. The modules could also be assembled into a physical structure, such as a cube or cuboid, in which virtual sound sources emit to the outside.

With the decentral structure, many very small sound transducers can be controlled. In the future, the application of MEMs [8] in conjunction with wave field systems could open up new perspectives. The integrated transducers can be built in large numbers on a common carrier base with other integrated components. In fact, this surface could then, like a curtain, generate wavefronts that are completely free of the aliasing effects that are more conventional in the relatively large elementary waves deviating from the theoretical approach of wave field synthesis

Sound transducers are inevitable.

On the support of the components would then again, analogous to the modular structure, to form groups of sound transducers that from a distributed system

To be supplied to control units. In the future, such microstructures could be used for a combined reproduction of auditory and visual information.

Detailed description of an embodiment

The device is shown in FIGS. 1 and 2. It will be explained with reference to these drawings.

Fig. 1 shows a modular device of sound transducers according to the principle of wave field synthesis (1). With her virtual sound sources (2) are shown, the position of which is given in a coordinate system with respect to the coordinate origin (3). The coordinate origin may be at the position of a listener in the playback room, but it may be set arbitrarily. In any case, the vector of a

Reference point (4) of the device may be known from sound transducers to this coordinate origin. Then the respective reference point in each of the modules (5) in the Device from Schallwandlem given by the placement of the module in the system and the edge length of the modules. In the module, the position of each individual transducer (6) is given to each individual transducer.

Because in the coordinate system then the positions of the virtual sound sources to coordinate origin (3) and all vectors of the reference points to each individual transducer are known, the position of each individual virtual sound source can be determined to each individual transducer by adding the individual vectors.

FIG. 2 shows that all the audio signals and data are routed to each module. This can be done via separate lines (1) and (2) or all information can also be transmitted via a common protocol to the modules. The amount of data is relatively small because only the position of the virtual sources in the coordinate system and their assignment to the audio signals must be transmitted. This allows an update of the positions in very short time intervals. Correspondingly quickly, the signals of all input sources delayed and added up from all according to the module position in the arrangement of sound transducers of all input sources can be fed to the corresponding power amplifier for the few transducers in the module.

In the devices described here from sound transducers according to the principle of wave field synthesis for the reproduction of sound events does not have on the

psychoacoustically related formation of phantom sound sources between speakers are used, but the sound field is physically reconstructed. From the audio signal itself (content) and data on its shape (form) Huygens principle wave fronts from elemental waves are reconstructed. It creates virtual sound sources that do not physically differ from the wavefront of the real sound source.

The audio signal will appear on the playback page in a renderer for each

Elementary wave folded into the spatial impulse response of the recording room [2].

For correct reproduction, the starting points of the elementary waves should be close together. The virtual sound sources can only arise in the area of the arrangement of sound transducers. Therefore, their number becomes very large when a two-dimensional sound transducer surface is built up.

Thus, the requirements for the renderer prior art systems increase, the control of a large number of transducers requires a high computational effort. In practice, the principle of wave field synthesis was therefore usually reduced to a horizontal transducer array. In these systems, the wave field synthesis is usually on the listener's horizontal plane is reduced, the third dimension is lost in the reproduction of the sound events.

With the solution according to the invention, however, the necessary computing power can be distributed decentrally, because the amount of data to be transmitted between the subsystems does not increase with the number of sound transducers. The system becomes freely scalable.

This even makes it possible to construct arbitrarily large, two-dimensional devices from sound transducers according to the principle of wave field synthesis, which can realize rapid changes in the position of the virtual sound sources with the associated Doppler effects without artifacts.

According to an embodiment of a decentralized device from

Sound transducers based on the principle of wave field synthesis, the synthesis of

Wavefronts from the audio signals and the associated data for the individual transducers contained within the respective assembly, wherein the geometric position of a reference point within the coordinate system for the model-based approach of wave field synthesis for each individual assembly by their placement in the arrangement of Schaliwandlern and the Edge length of the individual modules is determined and the position of each transducer in this

Coordinate system is defined by its arrangement to this reference point of the assembly, so that solely from the arrangement of the assemblies in the arrangement

Sound transducers can determine the position of each individual transducer in the coordinate system from the vector addition to the respective higher reference point.

According to a development, the assemblies are enclosed by a module housing or are formed of equal size segments in a structure of components.

Typically, the arrangement of transducers is freely scalable in size because the computing power in the central unit does not increase with the number of transducers in the system.

According to a further development, all the audio signals and the data for the synthesis of the wavefronts are fed to all assemblies of the device, wherein in each assembly, the data are processed, resulting from the position of the respective assembly within the array of sound transducers.

Typically, the position of the individual transducers within an assembly is stored relative to a fixed reference point of the assembly in the assembly.

According to a further development, the position of a fixed reference point of each module is determined to the position of a reference point of the device from sound transducers by informing the module at which position it is installed within the device of sound transducers and they are stored therefrom with the aid of Dimensions of the individual modules, which can also be designed as a module, the position of their reference point to the central reference point of the device

Can determine sound transducers.

According to yet another embodiment, the modules are different density with

Sound transducers fitted. Thus, the effort can be reduced in the reproductive areas, which are less important for the human perception of sound events.

The assemblies can be constructed in a closed plane, closed row.

But the assemblies can also be constructed so that they are not arranged in a closed plane or closed row.

According to a further development, the sound transducers are associated with sub-surfaces, which can form a body that can radiate the wavefronts in a common system in different directions.

According to a further development, a system for image reproduction is applied to the same carrier system that carries the sound transducer.

According to yet another development, the modules or modules are combined in pre-assembled units. This allows a faster setup of the system.

According to a further embodiment, a decentralized device constructed from acoustic transducers according to the principle of wave field synthesis on multiple modules, each having a plurality of sound transducers and an assembly control,

wherein each module controller is arranged to generate drive signals for the sound transducers of its assembly of audio signals and associated data for the form for the synthesis of the wavefronts.

This structure also enables a system for image reproduction to be applied to a support system carrying the sound transducers.

The features of the various embodiments described herein may also be combined.

Bibliography

[1] Berkhout, A.J. (1988): A holographic approach to acoustic control. Journal of the Audio

Engineering Society, Vol.36, No.12, December 1988, pp.977-995.

[2] http://www.hauptmikrofon.de/theile/WFS Theile VDT-Maqazin 2 2005.pdf

[DE] 10 2005 001 395 A1 [4] William Francis Wolcott IV: Wave Field Synthesis with Real-time Control

[5] Dipl. Ing (FH) Rene Rodigast, Frauenhofer - Institute for Digital Media Technology IDMT:

Voice reproduction or concert acoustics - Acoustic simulation in the sound system, 8.0 E 37 Messe Frankfurt / Prolight + Sound 2013

[61 http://iosono-sound.com/assets/files/IOSONO IPC100 brochure.pdf

[7] http: // wf s vnt h. so urcef orge. n et / thes is. pdf

[8] John J. Neumann, Jr. and Kaigham J. Gabriel, CMOS-MEMS Membrane for Audio-Frequency Acoustic Acuation, Electrical and Computer Engineering Dept., Carnegie Mellon University, 2001, p. 236-239, XP-002240602.

Claims

claims

1. Decentralized device of sound transducers according to the principle of

Wave field synthesis, comprising several assemblies, each one multiple

Have sound transducers,

wherein the device is set up by means of a model-based approach to perform a synthesis of the wavefronts of audio signals and the associated data on the form for the sound transducers contained in the assemblies within the respective assembly and to control the sound transducers within the respective assembly with drive signals corresponding to the synthesis.

2. The decentralized transducer device of claim 1, wherein the geometric position (6) of a reference point (5) of the assembly within a coordinate system for the model-based approach of wave field synthesis for each individual assembly by their placement in the array of transducers and the edge length of the individual assemblies is determined and / or the geometric position (6) of each transducer in this coordinate system is defined by its arrangement to this reference point (5) of the assembly, so that solely from the arrangement of the assemblies in the array of transducers the position of each individual transducer in the coordinate system from a vector addition to the respective parent reference point (6) can be determined.

3. Decentralized device of sound transducers according to one of

previous claims,

characterized,

that the assemblies are enclosed by a module housing and / or that they are formed from the same sized segments in a structure of components.

4. Decentralized device of sound transducers according to one of

previous claims,

characterized,

that the arrangement of sound transducers is freely scalable in size.

5. Decentralized device of sound transducers according to one of

previous claims,

characterized, that all audio signals and the associated data for the synthesis of

Wavefronts are led to all components of the device, wherein in each module, the data are processed, resulting from the position of the respective assembly within the array of sound transducers.

6. Decentralized device of sound transducers according to one of

previous claims,

characterized,

the position of the individual sound transducers within an assembly is stored relative to a fixed reference point of the assembly in the assembly.

7. Decentralized device of sound transducers according to one of

previous claims,

characterized,

that it can determine the position of a fixed reference point of each module to the position of a reference point of the device from sound transducers, which is communicated to the module at which position it is installed within the device from Schallwandlem and they are stored therefrom by means of

Dimensions of the individual modules, which can also be designed as a module that can determine the position of its reference point to the central reference point of the device from sound transducers.

8. Decentralized device of sound transducers according to one of

previous claims,

characterized,

that the modules are also equipped with different densities with sound transducers.

9. Decentralized device of sound transducers according to one of

previous claims,

characterized,

the assemblies can also be arranged in a non-closed plane or in a non-closed row.

10. Decentralized device of sound transducers according to one of

previous claims,

characterized, that the sound transducers can be assigned to sub-areas, which can emit the wavefronts in each case in a different direction. 1. Decentralized device comprising sound transducers according to one of the preceding claims,

characterized,

that the module or modules are combined in pre-assembled units.