JP4620468B2 - Audio reproduction system and method for reproducing an audio signal - Google Patents

Abstract

An audio playback system is divided into a central wavefield synthesis module (10) and a multitude of decentrally arranged loudspeaker modules (12a-12e). Synthesis signals for the individual loudspeakers and corresponding items of channel information, which are assigned to the synthesis signals, are calculated in the central wavefield synthesis module. The synthesis signals for a loudspeaker together with associated items of channel information are then transmitted to corresponding loudspeaker modules via a transmission link (16a-16e). Each loudspeaker module receives the synthesis signals and associated items of channel information that are intended for the loudspeaker assigned to the loudspeaker module. A decentralized audio rendering and digital-to-analog conversion takes place inside the loudspeaker modules in order to decentrally generate the actual analog loudspeaker signals in spatial proximity to each loudspeaker. The division into a central wavefield synthesis module and a multitude of decentralized loudspeaker modules enables the production of audio playback systems that can be scaled with regard to price in order to offer different size systems, which can be scaled in terms of price, for, in particular, cinema playback spaces that vary greatly in size.

Description

  The present invention relates to an audio playback system, and more particularly to an audio playback system suitable for use in a variable-size playback room such as a movie theater, and an audio playback system based on wave-field synthesis.

  In the field of consumer electronics, there is a growing demand for new technologies and innovative products, and an important prerequisite for the success of new multimedia systems is that they provide optimal functions and capabilities, respectively. . This is achieved by digital technology and in particular computer technology. Examples include devices that provide improved realistic audiovisual impressions. Conventional audio systems have significant weaknesses not only in the natural environment but also in the quality of spatial sound reproduction in a virtual environment.

  Methods of multichannel speaker reproduction of audio signals have been known and standardized for many years. All common technologies have the disadvantage that both the speaker location and the listener location are already incorporated into the transmission format. Audio quality is significantly degraded due to incorrect placement of speakers relative to the listener. Optimal sound is possible only in a small area of the reproduction room, the so-called sweet spot.

  An excellent natural spatial impression in audio reproduction as well as a powerful enclosure can be obtained with the help of new technology. The basis of this technology, so-called wavefront synthesis (WFS), was studied at Delft University of Technology and was first published in the late 80s (Non-patent Document 1).

  Since this method requires tremendous requirements for calculation operations and transmission rates, wavefront synthesis has been rarely used in practice. Today, advances in the field of microprocessor technology and audio coding are essential to enable the use of this technology in specific applications. The first commercialization is expected in the professional field next year. Within a few years, the first products using wavefront synthesis will appear in the consumer market.

  The basic concept of WFS is based on the application of Huygens principle on wave theory. That is, all points captured by the wave are the starting points of elementary waves that propagate in a spherical or circular shape.

  When applied to acoustics, any shape of incoming wave-front can be reproduced by a large number of speakers (so-called speaker arrays) arranged adjacent to each other. In the simplest case, when a single point source is played and the speakers are in a linear arrangement, the audio signal of each speaker with time delay and amplitude scaling is the emitted sound field of the individual speaker. fields) must be supplied to overlay properly. When there are multiple sound sources, the contribution to each speaker is calculated separately for each sound source and the resulting signals are added. If the sound source to be reproduced is in a room with a reflecting wall, the reflection must also be reproduced as an additional sound source via the speaker array. Therefore, the labor involved in the calculation strongly depends on the number of sound sources, the reflection characteristics of the recording room, and the number of speakers.

  The particular advantage of this technique is that it allows a natural spatial acoustic impression in a wide reproduction room. In contrast to known techniques, the direction and distance from the sound source is reproduced very accurately. To a limited extent, a virtual sound source can even be positioned between the actual speaker array and the listener.

  Wavefront synthesis works well when the surrounding conditions are known, but variation occurs when the conditions change and when wavefront synthesis is performed based on surrounding conditions that do not match the actual surrounding conditions.

  The ambient condition can also be expressed by a peripheral impulse response.

  This will be explained in more detail with respect to the following example. Assume that the speaker emits a sound source signal against a wall where reflection is not desired. In this simple example, room compensation by using wavefront synthesis would be as follows: That is, the reflection of this wall is first determined to determine when the acoustic signal reflected by the wall reaches the speaker again and how much amplitude the reflected acoustic signal has. If reflection from this wall is not desired, wavefront synthesis offers the possibility of removing reflection from this wall in the following manner. In other words, in addition to the original audio signal, a signal having a phase opposite to that of the reflected signal is given to the speaker having the corresponding amplitude, and as a result, the forward compensation wave (forward compensation wave) removes the reflected wave. The reflection from the above-mentioned wall in the periphery is removed. This method can be performed by first calculating the surrounding impulse responses and determining the state and position of the wall based on these surrounding impulse responses, at which time the wall reflects incident sound. Interpreted as a mirror source meaning sound source.

  When measuring the impulse response of these surroundings first, and then calculating the compensation signal to be applied to the speaker that overlays the audio signal, the removal of reflections from this wall is performed by listeners around these surroundings. With regard to sound, it is executed so as to have the impression that this wall does not exist at all.

  However, accurate determination of the chamber impulse response is essential for optimal compensation of reflected waves so that no over- or under-compensation is performed.

  Thus, wavefront synthesis allows for accurate mapping of virtual sound sources over a wide reproduction range. At the same time, wavefront synthesis provides recording engineers and acoustic engineers with new technical and creative potential for the design of composite acoustic scenes. Wavefront synthesis (also known as WFS or sound case) developed at the Delft University of Technology at the end of the 1980s represents a holographic method of sound reproduction. The Kirchhoff-Helmholtz integral is the basis for this. It means that any sound field within a closed volume can be generated via the distribution of monopole and dipole sound sources (speaker arrays) on the surface of this volume Is shown. Details of this can be found in Non-Patent Document 2 and Non-Patent Document 3.

  In wavefront synthesis, a synthesized signal is calculated for all speakers in the speaker array from an audio signal radiated by a virtual sound source at a virtual position. At this time, with respect to amplitude and phase, when the resultant wave from the superposition of sound waves output from individual speakers existing in the speaker array is a real sound source at the virtual position, this virtual sound source at the virtual position The synthesized signal is formed so as to coincide with the wave that will be generated from this virtual sound source.

  Usually, a plurality of virtual sound sources are present at different virtual positions. The calculation of the synthesized signal is executed for all virtual sound sources at all virtual positions, and usually a synthesized signal to a plurality of speakers is obtained from one virtual sound source. Therefore, when viewed from one speaker, the speaker receives a plurality of synthesized signals generated from different virtual sound sources. As a result of the superposition of these sound sources enabled by the linear superposition principle, a reproduction signal that is actually radiated by the loudspeaker is generated.

  The larger the speaker array, that is, the more individual speakers are provided, the better the possibility of wavefront synthesis. However, this also increases the computational operations that the wavefront synthesizer must perform, since typically channel information must also be taken into account. In other words, it basically means that there are individual transmission channels from all virtual sound sources to all speakers, and basically, synthetic signals for all speakers are generated from all virtual sound sources. This means that there may be cases where all the speakers obtain the same number of synthesized signals as the number of virtual sound sources.

  In particular, there is a possibility that a virtual sound source may move in the application to a movie theater, and the reason why the possibility of wavefront synthesis is reduced is as follows: calculation of composite signal, calculation of channel information, channel information and composite signal It can be mentioned that a large amount of calculation operation is required for the generation of the reproduction signal by the combination.

  Moreover, it should be noted here that the quality of audio reproduction increases with the number of speakers provided. This means that the more speakers that are present in the speaker array, the higher and realistic the audio playback quality will be.

  In the above scenario, a fully rendered and analog-digital converted playback signal for an individual speaker can be transmitted from the wavefront synthesis central unit to the individual speaker via, for example, a two-wire line. This has the advantage that no further means for synchronization purposes are required since it is almost guaranteed that all speakers operate in synchronism. On the other hand, however, the wavefront synthesis central unit can always be made only for reproduction by a specific reproduction room and a fixed number of speakers. This means that an individual wavefront synthesis central unit must be created for every regeneration chamber. In addition, the calculation of the audio reproduction signal must be executed at least partially in parallel and in real time, particularly with respect to a large number of speakers and a large number of virtual sound sources, so that a large amount of computing power must be provided. Means.

  Particularly with audio playback systems intended for movie theaters, there is the problem that the playback rooms of movie theaters are remarkably diverse with respect to their size. Some cinemas have very large movie screens and / or several smaller ones provided for movies that do not have as many spectators as movies shown on a large screen. Some have a movie screen. However, different movie theaters have different sized playback rooms, and the difference can be as much as 100 times, especially when audio playback is considered not only for movie theaters but also for concert halls, for example.

  In order to provide an audio playback system based on wavefront synthesis in such various audio playback rooms, for example, individual wavefront synthesis central units must be made for all the playback rooms, which would be individually produced So unacceptable in terms of price.

  On the other hand, it would be possible to construct a fully equipped wavefront synthesis central unit. That is, it is possible to cope with the number of connectable speakers, that is, the number of analog signal outputs. Inside, there is a device including an arithmetic processing unit designed for the maximum number of analog outputs, that is, the maximum number of connectable speakers. Conceivable.

  However, such a system would lead to an audio playback system for a small playback room having almost the same price as an audio playback system for a very large playback room, which is probably in a small playback room. It will not be accepted by the operator. For audio playback system providers, there is a particular need to focus on medium to small playback rooms. In other words, attention should be paid to “smallest” reproduction rooms such as private rooms or small restaurants and bars.

Therefore, the possibility of constructing the maximum unit described above is disadvantageous, and rapid market acceptance cannot be expected immediately.
(Berkhout, AJ; de Vries, D .; Vogel, P .: Acoustic Control by Wave-field Synthesis. JASA 93, 1993) MMBoone, ENG Verheijen, PF v. Tol, "Spatial Sound-Field Reproduction by Wave-Field Synthesis", Delft University of Technology Laboratory of Seismics and Acoustics, Journal of J. Audio Eng. Soc., Vol. 43, No. 12 , December 1995 Diemer de Vries, “Sound Reinforcement by Wavefield Synthesis: Adaption of the Synthesis Operator to the Loudspeaker Directivity Characteristics”, Delft University of Technology Laboratory of Seismics and Acoustics, Journal of J. Audio Eng. Soc., Vol. 44, No. 12 , December 1996

  An object of the present invention is to provide a basic concept of audio reproduction that is highly accepted by the market.

  This object is achieved by an audio reproduction system according to claim 1, a method for reproducing an audio signal according to claim 19, or a computer program according to claim 20.

  The present invention is based on the finding that audio playback systems that seek to achieve market acceptance must be scalable. However, this scaleability not only exerts on the computing power provided, but also has an impact on the price of the audio playback system. In other words, this means that an audio playback system for a large playback room can be more expensive than an audio playback system for a small playback room. In other words, an audio playback system for a small playback room must be significantly less expensive than an audio playback system for a large playback room.

  In the concept of maximum unit construction described above, the price difference depending on the scale was small. Because the price difference only occurs depending on the number of individual speakers, but because of the fact that many speakers are mass-produced and the new concept of integrating into a building with a reproduction room, such speakers are inexpensive. Because it can be provided.

  According to the present invention, the audio reproduction system is divided into a central wavefront synthesis module and a number of individual speaker modules connected in a distributed manner to the central wavefront synthesis module.

  The central wavefront synthesis module receives an audio signal having a plurality of audio tracks, calculates a composite signal on the one hand, and calculates channel information for a channel from the virtual position to the actual speaker position on the other hand.

  Further, the central wavefront synthesis module not only supplies one or more synthesized signals to be reproduced by each speaker to each speaker, but also a virtual position or virtual sound source serving as a source of the one or more synthesized signals. Are configured to provide channel information regarding audio channels from to each speaker. Here, empirically, it is very rare for all speakers to receive a composite signal whose energy content exceeds a certain threshold, so a valid data rate transmission limit has already been obtained. Has been. Thus, the central wavefront synthesis module of the present invention already has the option of supplying only the synthesized signal that is valid for the individual speakers and only the channel information for this synthesized signal to the distributed speaker module.

  The speaker module of the present invention is realized in a distributed manner, is directly connected to the speaker, and is preferably disposed spatially close to the speaker. All speaker modules are equipped with a receiver to receive one or more combined signals for each speaker and channel information associated with the combined signals. Further, all the speaker modules are provided with a rendering means, and calculate a reproduction signal of the speaker by using the synthesized signal and channel information related to the supplied synthesized signal. Further, each speaker module preferably has a signal processing means having one digital amplifier, another digital signal processing means, and a digital-to-analog conversion for generating an analog speaker signal supplied to each speaker as a reproduction signal. With a vessel. In order to connect the central wavefront synthesis module and the distributed speaker module, a plurality of transmission paths are provided, and each transmission path extends from the central wavefront synthesis module to the individual speakers.

  The rendering operation is very computationally intensive, which is particularly in the form of a DSP (digital signal processor) or wired circuit, especially when considering the multipliers provided for all individual speakers. Significantly contributes to the cost of the required circuit hardware. Preferably, the rendering means operates by using the channel impulse response as channel information and performs a computation time intensive convolution that can be performed directly in either the time domain or the frequency domain. At this time, conversion to the frequency domain and conversion from the frequency domain are necessary, and as a result, a great deal of processing is required together with an actual multiplication operation in the frequency domain. Here, it should be particularly noted that the rendering unit not only calculates one individual combined signal, but usually always calculates a large number of combined signals corresponding to the number of virtual sound sources.

  As a result of the concept of the present invention, the operations that can be performed in a distributed manner are shifted from the central wavefront synthesis module to the distributed speaker module, so that in the best case, all speakers have equivalent meanings. Only the central wavefront synthesis module is executed, and all computations relating to only one or more speakers connected to one speaker module are executed in a distributed manner in that speaker module.

  Thus, the cost of the central wavefront synthesis module can be significantly reduced, but on the other hand, the speaker module whose price is no longer negligible is used mainly for audio rendering operations performed on the speaker module. Incurs the cost of increased costs.

  However, the audio reproduction system of the present invention can be scaled in both performance and price. Since the central wavefront synthesis module for a large number of playback rooms may be offered at a low price, the overall system cost due to the cost of the central unit and the distributed speaker module will depend on the number of speakers installed and thus the playback. It depends greatly on the size of the room.

  In other words, large playback room operators still have to pay a certain price for a playback system for a large playback room, while small playback room operators have much lower audio playback systems. It becomes possible to purchase at a price. This is because the number of speakers, that is, the number of expensive and cost-intensive speaker modules, is significantly reduced compared to a large-scale reproduction room.

  Therefore, the audio reproduction system of the present invention can provide an audio reproduction system for a small reproduction room at a considerably reduced price as compared with a large reproduction room, and as a result, the audio / video component can be provided. You can expect to be welcomed in a very competitive market.

  In a preferred embodiment of the present invention, the central wavefront synthesis unit is formed to allow processing of motion picture film recorded in a conventional audio format for motion picture film. Here, the general recording format is, for example, 5.1 surround format, 7.1 format, or 10.2 format. In the case of the 5.1 format, such a motion picture film has six audio tracks, namely the “rear left”, “rear right”, “front left”, “front right”, “front center” channels as well as the subwoofer channel. Includes audio tracks for. In the audio reproduction system according to the present invention, the reproduction of such a movie film, which can be said to be conventional with respect to audio technology, is a virtual position at a virtual position that can be selected according to the preference of the acoustic engineer and the reproduction room operator This can be achieved by arranging an audio track as a sound source. Thus, the playback method of the present invention, which is scalable in terms of price and compatible in an audio playback system, is a movie having an audio track that is perfectly suited for wavefront synthesis with the meta information required for recording settings. / Even in the era when there is very little video film, it has the potential to spread audio playback systems based on wavefront synthesis at once.

Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.
It is a conceptual diagram of the audio reproduction system of this invention. It is a block diagram of the central wavefront synthesis module of the present invention. It is a block diagram of the distributed speaker module of this invention. FIG. 2 is a block diagram of a preferred embodiment of an audio rendering unit within a distributed speaker module. A basic representation of compatible playback with a large sweet spot. It is a basic diagram showing a state where a plurality of composite signals each having channel information are input to one speaker in order to obtain a reproduction signal of speaker LSi. It is a basic diagram showing a channel from a virtual sound source to a real speaker together with variables that can affect the channel.

  The playback system of the present invention is basically divided into two parts as shown in FIG. One part is the central wavefront synthesis module 10. The other part is composed of individual speaker modules 12a, 12b, 12c, 12d, 12e, which are connected to real physical speakers 14a, 14b, 14c, 14d, 14e as shown in FIG. It should be noted that the typical number of regions of the speakers 14a-14e ranges from 50 or more, more typically 100 or more. When individual speaker modules are provided for all speakers, a corresponding number of speaker modules are also required. Depending on the application method, it is preferable to address a small group of adjacent speakers from one speaker module. Here, for example, whether one speaker module connected to four speakers supplies the same reproduction signal to these four speakers, or different synthesized signals for these four speakers. It is not important whether or not is calculated. Such a speaker module is actually composed of a plurality of individual speaker modules, but is physically integrated into one unit.

  Individual transmission paths 16a to 16e exist between the wavefront synthesis module 10 and all the individual speaker modules 12a to 12e, and all the transmission paths are coupled to the central wavefront synthesis module and the individual speaker modules.

  As a data transmission mode for transmitting data from the wavefront synthesis module to the speaker module, a serial transmission format that provides a high data rate such as a so-called firewire transmission format or a USB data format is preferable. A data transmission rate in excess of 100 megabits per second is advantageous.

  The data stream transmitted from the wavefront synthesis module 10 to the speaker module is formatted in the wavefront synthesis module so as to correspond to the selected data format and is provided in a general serial data format. Is given. This synchronization information is extracted from the data stream by the individual speaker modules and is synchronized with respect to the digital / analog conversion to synchronize the individual speaker modules with respect to their playback, i.e. to obtain an analog speaker signal and resample it. Used to The central wavefront synthesis module operates as a master and all speaker modules operate as clients, where all individual data streams can obtain the same synchronization information from the central module 10 via different transmission paths 16a-16e. preferable. This ensures that all speaker modules operate synchronously, ie synchronized by the master 10. This is because, in the audio playback system of the present invention, in order to avoid loss of audio quality, that is, the synthesized signal calculated by the wavefront synthesis module is shifted in time to each individual speaker after each audio rendering. It is important to prevent output. The advantage of this concept is that the individual speaker modules do not need to be synchronized with each other. They all operate synchronously with the master, so they are automatically synchronized with each other. Connection between individual speaker modules is not desirable for the present invention. This is because the modular concept of speaker module scaling with respect to the size of the playback room requires simple addition of modules and does not have to achieve corresponding wiring between modules.

  FIG. 2 shows a block diagram of a central wavefront synthesis module according to a preferred embodiment of the present invention. First, the central wavefront synthesis module comprises input means 20 that is generally configured to receive an audio signal at the input, the audio signal having a plurality of audio tracks, and a sound source for all these audio tracks. Are associated with each other.

  Depending on the application method, the sound source position is an index indicating the position of the speaker relative to the listener in the reproduction room, and follows a standardized audio format such as 5.1 in order to obtain compatible reproduction. In this case, the audio signal has 5 + 1 = 6 audio tracks. Instead, the audio signal can have a greater number of audio tracks, which already exist as wavefront synthesis suitability signals and represent sound sources and audio objects at the actual recording location, respectively, which use wavefront synthesis Thus, the audio signal reproduction is mapped as a virtual sound source in the reproduction room.

  Furthermore, in a preferred embodiment of the invention, the input means 20 is used as a main control unit, which unit preferably has further functions. In particular, it has the function of a decoding module, as is commonly used in cinemas. Alternatively or additionally, the input means 20 may be formed as a DVD decoder that provides separate audio channels and audio tracks, respectively.

Alternatively, the input means 20 may be formed as an MPEG4 decoding module in which an audio track 21 intended for wavefront synthesis and corresponding sound source information 22 are already provided. In particular, the audio track 21 relates to the audio signal from the audio object in the recording setting, the position of the audio object in the recording setting, the characteristics of the audio object, in particular the characteristics relating to the size of the audio object, or the density of the audio object relative to the acoustic characteristics. To do.

  Furthermore, if necessary, it is preferable to transmit the characteristics of the recording room and the recording environment in addition to the audio track 21 and to take these characteristics into consideration in the wavefront synthesis. Information about each of the recording room and recording environment provides the listener with not only a visual representation but also an audible impression of the recording situation. In the reproduced sound, the spectator realizes whether the recording scene of the movie is outdoors or a small room such as a submarine. An outdoor recording scenario provides a relatively “dry” audio signal since there is little or no reflection in the recording environment, but the situation is quite different, for example, in a submarine. Here, the recording settings are each represented by a room with multiple reflections and an audio environment with multiple reflections. In this case, it is preferred that the audio track be recorded as dry as possible, i.e. without room acoustics in the recording room, and the room acoustics related to its characteristics be represented by additional meta information. This is because they can be transmitted in a standardized data stream according to standard MPEG4.

  The central wavefront synthesis module comprises means 24 for determining channel information on the one hand and wavefront synthesis signals for the individual speakers on the other hand. Furthermore, a means 25 for converting the sound source position 22 into a virtual position for wavefront synthesis is provided.

  Means 24 for determining audio channel information for all audio channels from the virtual position to the speaker position are individually formed, the virtual position depending on the sound source position (means 25) associated with the audio track, The audio channel information exists for all channels from all virtual positions to all speakers. Further, the means 24 is configured to calculate a composite signal from the virtual position to the loudspeaker by using the principle of wavefront synthesis described above and well known.

  Furthermore, the central wavefront synthesis module of FIG. 2 comprises means 26 for providing a synthesized signal to one or more speakers. The means 26 is configured to transmit channel information related to the synthesis information transmitted from the central wavefront synthesis module to the individual speaker modules via the respective transmission paths, and to perform audio rendering in those individual speaker modules. Yes. In some embodiments, it is preferable to transmit further information about the channel from the virtual position to the real speaker for each composite signal. This means that means 24 in the preferred embodiment of the present invention also provides channel information for all combined signals, interpolates it from the calculated channel information respectively, and means 26 the interpolated channel information. Means that the means 26 can start the transmission to the individual speaker module. Preferably, in order to save data transmission capacity, the means 26 is configured to eliminate invalid composite signals and transmit neither invalid composite signals nor their associated channel information. Therefore, an effective synthesized signal is transmitted from one virtual sound source to several speakers, while a synthesized signal can be similarly calculated for other speakers of the speaker array according to the theory of wavefront synthesis. Although they can, their signals are relatively small with respect to their performance within a certain period of time, and therefore often occur negligible due to reduced data transmission.

  Specifically, the means 24 includes functions used to preprocess the audio signal. Moreover, the means 24 controls the individual speaker modules, in particular the individual speaker modules introduce synchronization information directly or in connection with the data streams transmitted to them, so that all Are controlled so as to obtain central synchronization with respect to the central wavefront synthesis module.

  In the concept of the present invention, all processing operations common to all playback channels are performed by the central wavefront synthesis module, while different processing operations are performed in a distributed manner for each individual speaker and each individual playback channel. Is formed.

  Furthermore, the means 24 is configured to perform a simulation of wavefront synthesis information in the case of stereo signals, 5.1 signals, 7.2 signals, 10.2 signals, etc. for compatible playback. Therefore, the standard position of the speaker relative to the playback room for the standardized audio format is used as the sound source position.

  This will be described with reference to FIG. 5 shows not only the reproduction room 50 and the speaker array 52 extending around the reproduction room, but also a large number of virtual sound sources 53a to 53e arranged at virtual positions outside the reproduction room 50 as can be seen from FIG. Show. The means 24 is related to the means 25 of FIG. 1 and is formed for calculating the virtual position from the sound source information, ie for calculating the virtual position from the standard position index of the 5.2 signal that can be controlled manually. . In some embodiments, it is preferable to move the virtual position to, for example, an infinite region so that the speaker array 52 emits planar waves into the reproduction chamber 50. This has the result that the so-called sweet spot, i.e. the area in the reproduction room where an optimal acoustic impression is obtained, is significantly enlarged compared to the standard situation in which real 5.1 speakers are arranged in the reproduction room.

  Alternatively, a virtual sound source can be placed at a finite virtual position and modeled as a point sound source. This option has the advantage that the acoustic impression is more comfortable for the movie audience / listener. Plane waves have the property that the listener has the impression of sitting in a very large room, which leads to unpleasant perception, especially when the submarine scene is deployed on a screen, for example. In this connection, it should be noted that a typical motion picture film, for example with 5.1 audio tracks, does not contain information about the acoustic characteristics of the recording settings. Therefore, in such a case, it is preferable to find a compromise between a plane wave, that is, a virtual sound source at an infinite position or a virtual sound source at a finite position. In this regard, the audio playback system of the present invention further provides the ability to vary the virtual position of the virtual speakers 53a-53e depending on the scene of the movie. For example, if the scene is unfolded outdoors, the speakers can be placed in an infinite area. However, when the scene is developed in a small room, the speaker can be arranged closer to the reproduction room 50.

  In connection with compatible playback, in a preferred embodiment of the present invention, the input means 20 is configured to sample the audio track associated with the video signal by a specific time “delay” before the video signal. As a result, the sound associated with a certain time is sampled simultaneously with the video signal associated with a certain time in the individual speaker module after the processing in the wavefront synthesis module. In the audio playback system of the present invention, a negative “delay” must be measured so that sound and image are emitted together. If the negative delay is large, the signal can already be fully calculated, and the signal can be passed from the speaker module to the speaker in response to each synchronization signal, for example, to ensure synchronization of the image and sound. Can be output.

  In both the case of compatible reproduction and the case where the input signal includes already prepared wavefront synthesis information relating to the sound source in the recording setting, channel information calculation means 24 sends information relating to the reproduction room via line 27. As a result, it is preferred that the combined signal is processed using information about the playback room so that, for example, the acoustic properties of the playback room can be removed.

  Information about the playback room can be determined by the geometry of the playback room, or can be measured in the playback room by using speakers and a specific microphone array, where control and evaluation is controlled by the playback room. This can be done via the adaptation module 28. In one embodiment of the invention, the acoustic characteristics of the playback room are determined during playback, and the information about the playback room is reset accordingly, so that, for example, in the case of a full movie theater, optimal suppression of movie sound is performed. It is preferable to make it. At this time, it should be noted that the acoustic characteristics of the reproduction room are significantly different from those in the case where there are no people in the reproduction room, particularly in a small and full reproduction room.

  Furthermore, the adaptation module 28 for the reproduction room comprises a microphone array that can be used to measure the characteristics of the reproduction. Furthermore, the adaptation module 28 for the reproduction room comprises an algorithm for finding the position of the speaker array in the reproduction room. In addition, pre-processing of the measurement results is performed in order to perform optimal inverting of the reproduction room characteristics and speaker characteristics. At this time, the adaptation module 28 is preferably controlled by the means 24.

  In some embodiments, the adaptation module 28 may be needed only for the playback room system configuration. However, this adaptation module 28 can be used continuously during operation if continuous adaptation to the changing conditions of the regeneration chamber is desired.

  If the channel information calculation means 24 is used to process the WFS specific signal input to the means 20, additional WFS information, i.e. the characteristics of the audio object, the characteristics of the recording room, etc., are input audio. It is extracted from the signal and supplied to the means 24 via the WFS information line 29, so that this information can be taken into account in the channel information calculation.

  In this case, the central WFS module is further configured to perform preprocessing of the WFS processed audio signal. Furthermore, means 24 and / or means 26 are provided for obtaining synchronization between the picture and the sound, so that, as described above, a time code is preferably inserted in the serial data stream to the individual speaker module. Finally, as already explained, the channel information calculation means 24 controls the adaptation module 28 to control the measurement of the acoustic characteristics of the reproduction room, either before reproduction or during reproduction, if necessary. Play a role.

  The multiplexer / transmission stage 26 is configured to introduce synchronization information generated in either the means 24 or the control means 20 or the means 26 itself into the data stream to the speaker module. The speaker module is further supplied with a composite signal required for the individual speaker and necessary channel information.

  It should be further noted here that means 24 are further specified for calculating channel information and for calculating synchronization signals and for calculating dedicated composite signals and dedicated channel information for individual speakers. The speaker position in the playback room must be provided. This is shown symbolically by line 30 in FIG.

  Hereinafter, a preferred embodiment of the speaker module will be described with reference to FIG. First, the speaker module comprises a receiver / decoder block 31, which receives a data stream from the selection means, and from this data stream, combines signal 31a, associated channel information 31b, synchronization information 31c, To extract. The speaker module shown in FIG. 3 further comprises audio rendering means 32 as a central device, which means 32 by using one or more combined signals and by using channel information associated with these combined signals. Then, the reproduction signal for the speaker is calculated. Further, the speaker module includes signal processing means 33 with a digital / analog converter, and the signal processing means 33 generates an analog speaker signal supplied to each speaker LSi 34 in order to generate an audio signal. The synchronization information 31c extracted from the data stream by the receiver 31 is supplied to the signal processing means 33, in particular the resampler cooperating with the digital / analog converter. As a result, the loudspeaker is synchronized with the central wavefront synthesis module with the composite signal calculated by the means 24 of FIG. 2 superimposed on each other and with channel information attached, and thus all other loudspeaker modules. Can be output accurately in time so as to synchronize with each other.

  Therefore, the speaker module shown in FIG. 3 is characterized by a combination of a digital receiver, other signal processing means, and a digital / analog converter, and in particular, the signal processing means 33 can be provided with a digital amplifier. Alternatively, the signal can be amplified after digital / analog conversion, but digital amplification is preferred because of the more precise possibility of synchronization. Furthermore, it is preferable to couple the speaker 34 to the signal processing means 33 via a short analog line. However, if the line from the signal processing means 33 to the speaker 34 cannot be shortened, the synchronization is preferably performed on the digital side, so if the line length between the speaker module and the speaker is very different, Asynchronization can occur, which can lead to audible artifacts and loss of acoustic impression to be created by wavefront synthesis, so that each line of all speakers is either the same length or a given It is preferable that the length has a difference within an allowable range.

In a preferred embodiment of the present invention, the channel impulse response is transmitted as channel information in the time domain or frequency domain. In this case, the audio rendering means 32 is designed to perform a convolution of the individual synthesized signals and the channel information associated with these synthesized signals. This convolution can actually be realized as a convolution in the time domain, or can be performed in the frequency domain by multiplying the frequency domain analysis signal by a channel transfer function, if necessary. An embodiment optimized for processing effort is shown in FIG. FIG. 4 shows a preferred embodiment of the audio rendering means 32, time-frequency conversion blocks 34a, 34b, 34c for all the synthesized signals S _ji (t) and the conversion of the channel impulse response into the converted values of the synthesized signals. In order to multiply the value H _ji (f), multipliers 35a, 35b, 35c provided in all branches, an adder 36, and a terminal frequency-time conversion means 37 are provided, as shown in FIG. It is connected to the. In the configuration shown in FIG. 4, since synthesized signals each having a channel transfer function are added in the frequency domain, only a single frequency-time conversion means is provided for all speaker modules regardless of the number of synthesized signals. It is characterized by the fact that it only exists and the processing effort is reduced. Depending on the embodiment, the time-frequency conversion of the composite signal S _ji can be performed completely in parallel, or can be performed in series / parallel or completely in series if there is sufficient time. .

  As indicated so far, the preferred audio rendering means 32 shown in FIG. 4 is preferably a single frequency-to-time conversion implemented as an inverse FFT, irrespective of the number of synthesized signals supplied to the speaker module. Characterized by the fact that it has only means 37. In this case, the means 34a, 34b, 34c are realized as FFT (FFT = Fast Fourier Transform).

  The audio rendering means 32 shown in FIG. 3 is further configured to obtain special program information from the central wavefront synthesis module shown in FIG. Accordingly, the multiplexer / transmission stage 26 comprises a specific output for providing program information to the speaker module. Depending on the application, the program information can be multiplied with the composite signal and channel information in the data stream, but this is not essential.

Hereinafter, an example for transmitting the program information to the speaker module will be described. When channel information is expressed as a channel impulse response and transmitted to the individual speaker module, it is in the front region of the impulse response where the envelope has a higher amount than the threshold, rather than the entire impulse response, in the sense of saving data rate Preferably, only impulse response samples are transmitted. It should be noted here that the impulse response usually has a large value only for a short time, takes a small value in terms of acceleration, and eventually has a so-called “reverberation tail”. Although it is important for the acoustic impression, the sample is no longer high and its specific phase relationship is not strongly perceived by the ear. In this case, reverberant tails with envelopes below the threshold are preferably transmitted simply as envelope values rather than based on samples. In the present invention, the reverberant tail samples required by the audio rendering means 32 are generated by an audio rendering means that generates any sequence of 0's and 1's, the amplitude of which is weighted by the transmitted envelope value . For further data reduction, it is preferable to transmit only a small number of envelope values , interpolate between the envelope values , and then use the interpolated envelope values for weighting the random 0/1 sequence.

It should be noted that the random 0/1 sequence is preferably realized by “1” being a positive voltage value and “0” being a negative voltage value. The information about whether the audio rendering means has received channel information, which is a real sample up to a specific value and then is simply an envelope value , is shared between the transmitter and the receiver via the program information shown in FIG. There is no change.

  In addition, the wavefront synthesis module of the present invention includes a WFS mixing console not shown in FIG. 2, which includes an author system for generating a WFS sound representation.

Hereinafter, a procedure that forms the basis of the generation of the composite signal will be described with reference to FIG. Consider a system with three virtual sound sources at three virtual positions 60, 61, 62 and a speaker LSi 63 at a known real speaker position in the central WFS module. Furthermore, the virtual positions 60, 61, 62 of the virtual sound sources are derived using the sound source positions since they are supplied in the WFS processed input signal or by means 25 for calculating them. Thus, it is known to the central wavefront synthesis module. The combined signals S _1i , S _2i , S _3i are signals that the speaker 63 must radiate, and they are derived from the respective virtual positions 60, 61, 62. From this, it can be seen that all the speakers emit a superposition of a plurality of synthesized signals as described above.

In addition, channel j _i is defined between all virtual positions and all speakers, which can be described, for example, by an impulse response, transfer function, or any other channel information, as described in connection with FIG. Can be described. All desired characteristics can be incorporated into the channel description and then provide the channel information for each channel associated with the synthesized signal for the synthesized signal calculated by the wavefront synthesis module can do. If channel information is given as an impulse response describing the channel, convolution is applied. If the signal is in the frequency domain, multiplication is performed. Depending on the embodiment, alternative channel information may be used.

  Hereinafter, information that can affect the channel 70 from the virtual sound source 71 to the actual speaker 72 will be described with reference to FIG. First, the virtual position of the virtual sound source 71 is introduced in the channel information, that is, for example, in the channel impulse response. Furthermore, virtual sound source characteristics such as size and density are introduced. Thus, for example, a small triangle is represented and modeled differently than a large timpani. Furthermore, as shown in FIG. 7, the characteristics of the recording room are introduced into the channel transfer function. A further influencing factor is the system distortion of the entire audio playback system, which includes, for example, speaker distortion and malfunction of the speaker, respectively. Furthermore, information regarding the reproduction room is introduced into the channel information in order to achieve compensation of the acoustic characteristics of the reproduction room. From this information, for example, if it is known that the playback room has a wall facing the front of the speaker and reflections from that wall must be suppressed, each speaker is controlled taking this information into account. , So that these speakers contain signals that are 180 degrees phase shifted with respect to the reflected signal and have their respective amplitudes, so that the reflection is removed and the walls are acoustically transparent, i.e. to the listener Identification by reflection is no longer possible.

  Finally, the channel information can also be used to set a specific target playback sound. Therefore, the channel information is first generated by suppressing the sound of the playback room in the form of playback room compensation, and then the generated channel information is provided to the wavefront synthesis module, so that any other playback room It is preferable that sound can be simulated in one reproduction room.

  Depending on the conditions, the inventive method for reproducing an audio signal can be implemented in hardware or software. Implementation methods include digital memory media that contain electronically readable control signals, in particular discs or CDs, that can work with a computer system that can be programmed to perform the method described above. It is. In general, the present invention also resides in a computer program product having program code for performing the methods of the present invention stored on a machine readable medium when the computer program product is executed on a computer. In other words, the present invention can also be realized as a computer program having a program code for executing a certain method when the computer program is executed by a computer.

Claims (14)

A system for a plurality of speakers to reproduce the audio signal in placed the reproduction room at a predetermined loudspeaker position of the reproduction room, the audio signal has a plurality of audio tracks, each sound source position in each audio track In the audio playback system associated with
Input means for inputting the audio signal; means for mapping the sound source position associated with the audio track as a virtual position in the playback room; and all audio channels from all the virtual positions to all the speakers. Means for determining audio channel information indicating the acoustic characteristics of the audio signal, and means for calculating a synthesized signal to be reproduced by the speaker from the audio signal using the principle of wavefront synthesis, which are output from the individual speakers. The reproduction is performed so that the wave generated from the superimposition of the sound wave coincides with the wave that will be generated from the virtual sound source when the virtual sound source at the virtual position is a real sound source at the real position in terms of amplitude and phase. means for calculating a combined signal which does not include the acoustic properties of the chamber, to be reproduced by said respective loudspeakers One or a plurality of central wave-field synthesis module and means for supplying and the channel information associated with the combined signal and the combined signal,
A plurality of speaker modules, wherein one speaker module is connected to at least one speaker, and each speaker module is connected to the one or more speakers for each speaker supplied from the central wavefront synthesis module; A receiver for receiving a composite signal and the channel information; a rendering means for calculating a reproduction signal for the speaker by using the one or more composite signals and the channel information; and A plurality of speaker modules, comprising: signal processing means for generating analog speaker signals to be supplied to each of the speakers by a reproduction signal;
A plurality of transmission lines that are individually coupled to the central wavefront synthesis module and the other individually coupled to the speaker module, extending from the central wavefront synthesis module to all the speaker modules;
An audio playback system comprising: Each speaker module is coupled to the speaker associated therewith so that a spatial distance between the speaker and the speaker module is shorter than a spatial distance between the speaker module and the central wavefront synthesis module. The audio playback system according to claim 1, wherein: The audio channel information, an audio reproduction system according to claim 1 or 2, characterized in that it comprises an impulse response representative of the acoustic characteristics associated with the audio channels. The rendering means for calculating the reproduction signal comprises convolution means for performing one or more convolutions by using one or more of the combined signals with respective impulse responses. Item 4. The audio playback system according to Item 3. The rendering means includes
Time domain-frequency domain converting means for converting each of the synthesized signals from the time domain to the frequency domain ;
Multiplying means for multiplying the channel transfer function obtained by converting the respective channel impulse responses to the frequency conversion values of the respective synthesized signals;
Summing means for summing the composite signal multiplied by the channel transfer function ;
A single frequency domain-time domain transforming means for transforming the total signal into the time domain to obtain the reproduced signal;
The audio reproduction system according to claim 4, further comprising: 2. The audio reproduction system according to claim 1, wherein the signal processing means in the speaker module includes a digital amplifier. The audio track is part of a video or movie film, and the central wavefront synthesis module is configured to sample the audio track of the video or movie film shifted by a certain time before video playback, and the time , taking into account the processing time in the central wave-field synthesis and in the above speaker module module, any one of claims 1 to 6, characterized in that simultaneous reproduction of image and audio are selected so as to obtain The audio playback system described in 1. The audio signal is an audio signal of an audio object in the recording environment as the audio track, the position of the audio object in the recording environment, one or more characteristics of the audio object such as size or density, and / or the recording. audio reproduction system as claimed in any one of claims 1 to 7, characterized in that it comprises information about acoustic characteristics of the environment. 9. The audio reproduction system according to claim 8 , wherein the central wavefront synthesis module is configured to determine the virtual position from the position of the audio object in the recording environment. The central wavefront synthesis module determines the channel information using information related to the acoustic characteristics of the reproduction chamber so that acoustic effects of the reproduction chamber are reduced when sound waves are reproduced by the plurality of speakers. to the audio playback system according to any one of claims 1 to 9, characterized in that. The central wavefront synthesis module is
Before reproducing the audio signal, calculate a plurality of room impulse responses between the speaker and the microphone arranged in the reproduction room,
By interpolating the total impulse response of the reproduction chamber from the plurality of room impulse responses, and considering the total impulse response when calculating the channel information, by reducing the acoustic characteristics of the reproduction chamber,
The audio reproduction system according to any one of claims 1 to 10 , wherein the audio reproduction system is configured to perform adaptation to an acoustic effect of the reproduction room. The central wavefront synthesis module is configured to generate synchronization information and incorporate the synchronization information into a data stream to the speaker module, and the plurality of speaker modules receive the synchronization information from the central wavefront synthesis module. and audio reproduction according to the speaker module using it for synchronization to any one of claims 1 to 11, characterized in that it is formed so as to synchronize with the central wave-field synthesis module system. A method for a plurality of speakers to reproduce the audio signal in placed the reproduction room at a predetermined loudspeaker position of the reproduction room, the audio signal has a plurality of audio tracks, each sound source position in each audio track In the method associated with
Inputting the audio signal;
Mapping the sound source location associated with the audio track as a virtual location in the playback room ;
Centrally determining audio channel information indicating acoustic characteristics of all audio channels from all the virtual locations to all the speakers;
A step of calculating a synthesized signal to be reproduced by the speaker from the audio signal in the center using the principle of wavefront synthesis, wherein a wave generated by superposition of sound waves output from the individual speakers is located at the virtual position. When a virtual sound source is a real sound source at a real position, a composite signal that does not include the acoustic characteristics of the reproduction room is calculated so that the amplitude and phase coincide with the wave that will be generated from this virtual sound source. Steps,
Transmitting one or more of the combined signals and channel information associated with the combined signals to a plurality of speaker modules;
Using the one or more composite signals for each speaker and the associated channel information to calculate the speaker reproduction signal in a distributed manner;
Performing signal processing by using digital / analog conversion to generate an analog speaker signal;
Collectively extracting the analog speaker signal from the plurality of speakers;
A method comprising the steps of: A computer program used as program code for executing the method of claim 13 when the program is executed on a computer.

Images