JP2006109343A - Speaker array system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a speaker array system which freely generates an acoustic beam having an arbitrary directivity and an arbitrary direction, and is suitably used in an arbitrary space. <P>SOLUTION: Many speaker units SP are arranged on the external side surface of a cylindrical speaker array 300. A CPU 101 selects a plurality of speaker units to be used for output of the acoustic beam on the basis of the information for designating the directivity and direction of the acoustic beam to be outputted from the array section 300, and computes the delay time of a plurality of delay audio signal to be supplied to these speaker units. A DSP 201 performs signal processing for acquiring the plurality of delay audio signals from a common audio signal. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a speaker array system that outputs acoustic beams from a plurality of speaker units arranged in an array.

A speaker system in which a large number of speaker units are arranged is called a “speaker array”. In the speaker array, by controlling the delay and gain applied to the audio signal supplied to each speaker unit, the directivity of the speaker can be sharpened and the direction of the acoustic beam can be controlled. When the directivity is increased, the same energy is radiated to a narrower range, so the distance attenuation of the sound pressure is reduced, the sound can be heard clearly far, and the acoustic radiation in unnecessary directions is suppressed. be able to. Further, if the direction of the acoustic beam can be controlled, there is an advantage that restrictions on the speaker installation method are reduced because the speaker does not necessarily have to be directed to the sound listening position. This type of speaker array is disclosed in, for example, Patent Document 1.
JP-A-9-233591

  By the way, the conventional speaker array described above assumes a space in which the speaker array is used, and sets and uses delays and gains given to audio signals supplied to individual speaker units in advance so that preferable directivity and direction can be obtained. However, it was not suitable for use in an arbitrary space.

  The present invention has been made in view of the circumstances described above, and can freely generate an acoustic beam having arbitrary directivity and direction, and is suitable for use in an arbitrary space. The purpose is to provide.

The present invention is based on a speaker array composed of a plurality of speaker units arranged to form a convex surface, and information for designating the directivity and direction of the acoustic beam output from the speaker array, and outputting the acoustic beam in the speaker array. Selecting a plurality of speaker units to be used for calculating the delay times of a plurality of delayed audio signals supplied to the speaker units, and signal processing for obtaining the plurality of delayed audio signals from a common audio signal There is provided a speaker array system comprising a signal processing means for performing.
According to such a speaker array system, it is possible to radiate an acoustic beam in an arbitrary direction facing the convex surface on which the speaker unit is arranged and to control the directivity thereof.
In a preferred embodiment, the speaker array has a cylindrical shape.
When the speaker array is a cylindrical body, it is possible to radiate an acoustic beam in all directions of 360 degrees facing the outer surface of the cylindrical body. In addition to the cylindrical body, the shape of the speaker array may be a cone, a polyhedron, or a sphere.
In a preferred aspect, when the information specifying the directivity and direction of a plurality of types of acoustic beams generated from a plurality of types of audio signals is given, the calculation means includes the plurality of delayed audio signals for each acoustic beam. Signal delay time is calculated, and the signal processing means performs signal processing for obtaining a plurality of delayed audio signals for outputting each of the acoustic beams from the speaker array, from each of the plurality of audio signals.
According to this aspect, a plurality of types of audio signals can be emitted as a plurality of types of acoustic beams having arbitrary directivity and direction.
In another preferred embodiment, the signal processing means includes a band dividing means for dividing the audio signal into audio signals of a plurality of frequency bands, and a plurality of audio signals from each of the plurality of bands of audio signals obtained by the band dividing means. Delaying means for generating the delayed audio signal, and windowing means for multiplying the plurality of audio signals in the plurality of bands by a window function corresponding to each band.
According to this aspect, since the windowing process suitable for each band is performed for each frequency band, it is possible to effectively attenuate the wraparound sound generated outside the acoustic beam.
In another preferable aspect, the speaker array system uses a speaker unit that is not selected to be used for the output of the acoustic beam in the speaker array, and cancels the canceling sound that cancels the sneak sound generated outside the acoustic beam. A canceling sound generating means for generating is provided.
According to this aspect, it is possible to suppress the generation of the sneaking sound outside the acoustic beam.

Embodiments of the present invention will be described below with reference to the drawings.
<First Embodiment>
FIG. 1 is a block diagram showing an electrical configuration of the speaker array system according to the first embodiment of the present invention. As shown in this figure, the speaker array system includes a CPU board 100, a DSP board 200, and a cylindrical speaker array unit 300.

  FIG. 2 is a plan view of the cylindrical speaker array unit 300, and FIG. 3 is a side view of the cylindrical speaker array unit 300. As shown in these drawings, the cylindrical speaker array unit 300 is formed by arranging a large number of speaker units SP on the outer wall of a cylindrical body having a hollow portion. More specifically, the cylindrical speaker array unit 300 includes a plurality of line array units LU each having a plurality of speaker units SP. Each line array unit LU is a boundary between the first plate portion 301 forming a part of the outer wall of the cylindrical speaker array unit 300 and the adjacent line array unit LU (left adjacent in the example shown in FIG. 2). And a second plate portion 302 which is a wall. Here, the first plate portion 301 is a long plate material, and a plurality of speaker units SP are arranged here so as to form one line. The arrangement direction (line direction) of the speaker units SP is parallel to the axis direction of the cylindrical speaker array unit 300. The angle formed by the first plate portion 301 and the second plate portion 302 is determined by the number of line array units LU constituting the cylindrical speaker array portion 300.

  In the second plate portion 302 in each line array unit LU, as means for driving the speaker unit SP for one line provided in the line array unit LU, an interface board 310 and each speaker unit for one line are provided. A plurality of digital amplifier boards 320 corresponding to the SP are attached. The cylindrical speaker array unit 300 includes a power source 330 for supplying power to the circuits mounted on the digital amplifier boards 320, and the CPU board 100 and the DSP board 200 in FIG. 1 (both not shown in FIG. 2). ) Is fixed.

  In the present embodiment, the cylindrical speaker array unit 300 is provided as a means for generating an acoustic beam having an arbitrary directivity and direction. Here, with reference to FIG. 4 to FIG. 9, the generation principle of the acoustic beam in the present embodiment will be described. Of these drawings, FIGS. 4, 6 and 8 show a state where the cylindrical speaker array unit 300 is viewed from above, and FIGS. 5, 7 and 9 show the cylindrical speaker array unit. The state which looked at 300 from the side is shown.

  In the example shown in FIGS. 4 and 5, a convergent acoustic beam is generated so that the sound pressure is locally increased at a position P away from the cylindrical speaker array unit 300. If a plurality of sound waves having the same waveform and the same phase are simultaneously output from a plurality of virtual sound source positions on a spherical surface having a radius r centered on the position P, each sound wave is added in the same phase at the position P. The sound pressure at P increases locally. In the example shown in FIGS. 4 and 5, an acoustic beam that locally increases the sound pressure at the position P is generated by the cylindrical speaker array unit 300 by utilizing this fact. That is, it is as follows.

  First, in the present embodiment, a plurality of appropriate speaker units SP are selected from the cylindrical speaker array unit 300. In this selection process, for example, a plurality of speaker units SP that may be considered to be able to cause effective radiated sound to reach the position P, such as speaker units SP within a predetermined range centered on the speaker unit SP closest to the position P. Selected.

  Next, for each of the selected speaker units SP, the distance between the speaker unit SP and the position P is obtained, and the time required for the sound to propagate through the radius r and the difference ΔL between the distances is obtained. Then, the same audio signal is subjected to a plurality of types of delay processing for delaying the signal for each required time determined for each speaker unit SP to generate a plurality of delayed audio signals, and each delayed audio signal is assigned to the corresponding speaker unit. It is supplied to the SP. If it does in this way, the sound emitted from each speaker unit SP will be added in the same phase in the position P as mentioned above.

  In the examples shown in FIGS. 6 and 7, parallel acoustic beams that are neither converged nor divergent are generated. In order to generate such parallel acoustic beams, the following processing is performed in this embodiment. First, a speaker unit SP whose sound beam propagation direction is directly in front is obtained, and a plurality of speaker units SP within a predetermined range centering on the speaker unit SP are selected as speaker units to be operated for output of the acoustic beam. Here, the size of the range of the speaker unit SP to be selected may be determined in accordance with the size W of the radiation range of the acoustic beam in the horizontal direction. Next, a virtual plane orthogonal to the propagation direction of the acoustic beam is assumed behind the selected plurality of speaker units SP. Next, for each speaker unit SP, the distance ΔL from the speaker unit SP to the virtual plane is obtained, and the time required for the sound to propagate through this distance ΔL is obtained. Then, a plurality of delayed audio signals are generated by delaying the original audio signal for each required time determined for each speaker unit SP, and supplied to each speaker unit SP.

The example shown in FIGS. 8 and 9 is opposite to the example shown in FIGS. In this example, a divergent acoustic beam that expands the radiation surface according to propagation from each speaker unit SP is generated, and the effect of making it listen as if the sound source position is at the position P is realized. In order to realize this effect, in the present embodiment, first, a plurality of speaker units SP suitable for radiation in such a radiation direction are selected based on the radiation direction of the acoustic beam. Next, an imaginary spherical surface having a radius r centered on the position P is assumed behind the plurality of selected speaker units SP. Next, for each speaker unit SP, a distance ΔL from the speaker unit SP to the virtual spherical surface is obtained, and a time required for sound to propagate through this distance ΔL is obtained. Then, a plurality of delayed audio signals are generated by delaying the original audio signal for each required time determined for each speaker unit SP, and supplied to each speaker unit SP.
The above is the principle of generation of the acoustic beam in the present embodiment.

  In FIG. 1, a CPU 101 and a nonvolatile memory 102 are mounted on a CPU board 100. The CPU 101 includes a communication function for communicating with a device connected via a USB interface, a device connected via Ethernet (registered trademark), a device connected via a MIDI cable, and a host computer. ing. With this communication function, the CPU 101 receives information indicating the directivity and direction of an acoustic beam from a host computer or the like, and outputs an acoustic beam having such directivity and direction in accordance with the above principle. Parameters necessary for controlling the directivity and direction of the acoustic beam, such as the range of the speaker unit SP to be operated for the output of the sound and the delay time for obtaining a delayed audio signal for each speaker unit, are calculated and the DSP board 200 is operated. Supply. The details of the arithmetic processing performed by the CPU 101 will be clarified in the description of the operation of the present embodiment in order to avoid duplication of explanation.

  The nonvolatile memory 102 stores the ID of the CPU 101. Such an ID is stored because, as one embodiment of the present invention, a plurality of speaker array systems shown in FIG. 1 may be connected to a host computer that controls the entire system. Note that a DIP switch may be provided instead of the nonvolatile memory 102. FIG. 10 shows an example in which a plurality of speaker array systems shown in FIG. 1 are connected. In this example, cylindrical speaker array units 300 of a plurality of speaker array systems are vertically stacked to form a cylindrical speaker array unit that is long in the axial direction, and the CPU 101 of each speaker array system is connected to a host computer via Ethernet. It is connected. This configuration has an advantage that the directivity can be sharpened and the control can be performed down to a low frequency because a line array unit having a long line length is configured. In this aspect, the host computer transmits the ID of the communication partner in order to communicate with the CPU 101 of a certain system. Each CPU 101 compares the ID transmitted from the host computer with the ID stored in the non-volatile memory 102, and communicates with the host computer if they match.

  In FIG. 1, the DSP board 200 includes a DSP 201, a DIR (digital interface receiver) 202, a format conversion unit 203, a plurality of drivers 204, and a driver 205. The DSP 201 receives digital audio signals via the DIR 202, and processes these digital audio signals under the control of the CPU 101. Signal processing performed by the DSP 201 includes windowing processing in addition to the generation of the delayed audio signal described above. By executing these processes, the DSP 201 generates and outputs digital audio signals addressed to a plurality of speaker units SP that operate to output an acoustic beam. Note that details of the signal processing performed by the DSP 201 are also clarified in the description of the operation of the present embodiment in order to avoid duplication of explanation.

  The plurality of drivers 204 are respectively associated with the plurality of speaker units SP provided in the cylindrical speaker array unit 300. The digital audio signal addressed to a specific speaker unit SP output from the DSP 201 is converted into a format suitable for transmission by the format conversion unit 203, and then the one corresponding to the speaker unit SP that is the destination among a plurality of drivers 204 is obtained. Via the interface board 310.

  The DIR 202 generates a synchronization signal such as a clock based on the input digital audio signal. The generated synchronization signal is input to the DSP 201 and the format conversion unit 203 and supplied to all the interface boards 310 in the cylindrical speaker array unit 300 via the driver 205.

  One interface board 310 is provided for each of the plurality of line array units LU constituting the cylindrical speaker array unit 300. The interface board 310 provided in each line array unit LU is provided with a plurality of receivers 311 for receiving digital audio signals addressed to the plurality of speaker units SP provided in the line array unit.

Each line array unit LU is provided with a plurality of digital amplifier boards 320 on which circuits for driving a plurality of speaker units SP provided in the line array unit are mounted. The digital amplifier board 320 corresponding to one speaker unit SP receives a digital audio signal addressed to the speaker unit SP from the corresponding receiver 311 and converts it into an analog audio signal, and this analog audio signal. And an amplifier 322 that amplifies the signal and sends it to the speaker unit SP.
The above is the details of the configuration of the speaker array system according to the present embodiment.

  Next, the operation of this embodiment will be described. In the following, for convenience of explanation, a cylindrical speaker is represented by an index j for specifying a position in the arrangement direction of the line array units LU and an index i for specifying a position in the arrangement direction of the speaker units SP in the line array unit LU. Individual speaker units SPij provided in the array unit 300 are specified. 2 and 3, the cylindrical speaker array unit 300 has 36 line array units LU, and each line array unit LU has 12 speaker units SP. In addition, all speaker units constituting the cylindrical speaker array unit 300 are represented as SPij (i = 1 to 12, j = 1 to 36).

  FIG. 11 is a diagram for explaining processing contents of the CPU 101. The CPU 101 receives information indicating the directivity and direction of the acoustic beam to be generated from a host computer or the like. When generating an acoustic beam as shown in FIGS. 4 and 5, the information indicating the directivity and direction includes information indicating the coordinates of the position P where the sound pressure is locally increased. When generating parallel acoustic beams as shown in FIGS. 6 and 7, the information indicating the directivity and direction includes information indicating the propagation direction of the acoustic beam and the radiation width W thereof. 8 and 9, when generating an acoustic beam that allows the virtual sound source position to be heard at the position P, information indicating the directivity and direction is information indicating the coordinates of the virtual sound source position P. including. When receiving information indicating such directivity and direction, the CPU 101 selects the speaker unit SP to be operated for generating the acoustic beam. As shown in FIG. 11, when generating an acoustic beam that increases the sound pressure locally at a position P outside the cylindrical speaker array unit 300, the position P is positioned directly in front in the selection process of the speaker unit SP. An index jm of the line array unit LU is obtained, and a predetermined range ja (= jm−Δj) to jb (= jm + Δj) around this jm is obtained. The speaker unit SPij (i = 1 to 12, j = ja to jb) is a speaker unit that operates to generate an acoustic beam.

  As already described with reference to FIGS. 4 to 9, in this embodiment, in addition to generating an acoustic beam that increases the sound pressure at a specific position P, a parallel acoustic beam is generated (FIGS. 6 and 6). 7), or generate an acoustic beam that makes it sound as if the sound source position is at a specific position P (FIGS. 8 and 9). Since the example of the selection method of the speaker unit SP operated to generate each acoustic beam has already been described with reference to FIGS. 4 to 9, description thereof is omitted here. Regardless of which acoustic beam is generated, in the speaker unit SP selection process, the range ja to jb of the index j of the speaker unit SPij to be operated for generating the acoustic beam is determined.

  When the speaker unit SPij (i = 1 to 12, j = ja to jb) to be operated for generating the acoustic beam is determined, the CPU 101 delays Dij (i = 1) of the delayed audio signal supplied to these speaker units. ˜12, j = ja to jb). This calculation method has already been described with reference to FIGS.

  Next, the CPU 101 uses four types of window function values WAij (i = 1 to 12, j = ja to jb), WBij (i = 1 to 12, j = ja to jb), WCij (i = 1 to 12, j). = Ja to jb) and WDij (i = 1 to 12, j = ja to jb), delay time Dij (i = 1 to 12, j = ja to jb), and speaker unit SPij (to which these should be applied) i = 1 to 12, j = ja to jb) are associated with each other and supplied to the DSP 201. Hereinafter, these pieces of information are collectively referred to as acoustic beam configuration parameters.

The window function values WAij (i = 1 to 12, j = ja to jb) and the like are multiplied by the delayed audio signal to suppress side lobes that leak outside the intended radiation range of the acoustic beam. The value is determined so as to suppress the level of the delayed audio signal supplied to the peripheral speaker units among the speaker units SPij (i = 1 to 12, j = ja to jb) that are operated for generating the sound. In the present embodiment, considering that the window function effective for suppressing the side lobe depends on the band of the audio signal, four types of window functions WA, WB, WC corresponding to each band obtained by dividing the audio frequency band into four, WD is prepared.
In the present embodiment, a plurality of acoustic beams may be generated by the cylindrical speaker array unit 300 based on a plurality of digital audio signals. In this case, information specifying the directivity and direction of each acoustic beam is given to the CPU 101 from, for example, a host computer. At this time, the CPU 101 obtains the acoustic beam configuration parameter for each of the plurality of acoustic beams, and sends it to the DSP 201.

  In the present embodiment, information indicating the directivity and direction of a certain acoustic beam may change with time. In that case, the CPU 101 obtains an acoustic beam configuration parameter corresponding to the newly instructed directivity and direction each time the instructed directivity and direction changes, and sends the acoustic beam configuration parameter to the DSP 201.

  As an aspect in which the information indicating the directivity and direction of the acoustic beam is changed with time, time-series information indicating the directivity and direction of the acoustic beam is recorded on, for example, a recording medium, and this is recorded by a sequencer. There may be a mode in which the data is read and supplied to the CPU 101. As a mode other than this, a mode in which the direction of the acoustic beam is instructed to the CPU 101 by operating a joystick, a rotary encoder, a 3D input device, or the like can be considered.

  FIG. 12 is a block diagram showing the signal processing performed by the DSP 201 in terms of hardware. In the example shown in FIG. 12, the DSP 201 is provided with arithmetic processing means for generating a delayed audio signal for generating an acoustic beam for each of the five types of digital audio signals S1 to S5. Five types of digital audio signals S <b> 1 to S <b> 5 are supplied via the DIR 202.

  In the following, among the five types of arithmetic processing means corresponding to the digital audio signals S1 to S5, arithmetic processing means corresponding to the illustrated digital audio signal S1 will be described, but the other digital audio signals S2 to S5 are described. The same applies to the arithmetic processing means corresponding to the above.

  The DSP 201 performs time alignment delay processing 2011 on the input digital audio S1. This is a process for adjusting the phase relationship between the digital audio signals S1 to S5. The delay time in the delay process 2011 is designated by the CPU 101.

  Next, the DSP 201 selects an LPF process 2012A that selects a signal in the lowest band among the bands obtained by dividing the audio frequency band into four parts, a first BPF process 2012B that selects a signal in the second band, and a third band that selects a third band. The 2BPF processing 2012C and the HPF processing 2012D for selecting the highest frequency signal are respectively applied to the digital audio signals that have been subjected to the delay processing 2011.

  Next, the DSP 201 performs an oversampling process 2013 on each digital audio signal that has passed through the LPF process 2012A, the first BPF process 2012B, the second BPF process 2012C, and the HPF process 2012D, and then performs a delay process 2014. The oversampling process 2013 is performed in order to increase the resolution of the delay time in the delay process 2014.

  In each delay processing 2014, the information regarding the delay time in the acoustic beam configuration parameter given from the CPU 101 for the digital audio signal S1 is referred to, and the oversampled digital audio signal is converted into the delay time Dij (i = 1 to 12). , J = ja to jb), a plurality of delayed digital audio signals are generated. These are delayed digital audio signals corresponding to the speaker units SPij (i = 1 to 12, j = ja to jb).

  Next, the DSP 201 performs windowing processing 2015A, 2015B, 2015C, and 2015D on the delayed digital audio signal obtained by each delay processing 2014, respectively. Here, in the windowing process 2015A, window function values WAij (i = 1 to 12, j = ja to jb) are extracted from the acoustic beam configuration parameters given from the CPU 101 for the digital audio signal S1. Then, the delayed digital audio signals corresponding to the speaker units SPij (i = 1 to 12, j = ja to jb) are respectively multiplied by the window function values WAij (i = 1 to 12, j = ja to jb). The same applies to the other windowing processes. In the windowing process 2015B, the window function values WBij (i = 1 to 12, j = ja to jb) in the acoustic beam configuration parameters are used. In the windowing process 2015C, the window function values are set. WCij (i = 1 to 12, j = ja to jb) is used, and window function value WDij (i = 1 to 12, j = ja to jb) is used in the windowing process 2015D.

  In parallel with the above processing, the DSP 201 performs a plurality of additions for synthesizing a plurality of digital audio signals to be supplied to all the speaker units SPij (i = 1 to 12, j = 1 to 36) of the cylindrical speaker array unit 300. Process 2017 is executed in parallel. Then, the DSP 201 executes a mapping process 2016 that delivers the delayed digital audio signal to be sent to each speaker unit SPij obtained by the windowing process to the addition process 2017 for each speaker unit SPij.

  The results of a plurality of addition processes 2017 corresponding to the speaker units SPij (i = 1 to 12, j = 1 to 36) are output from the DSP 201 via a plurality of addition processes 2018, and the format converter 203 and driver in FIG. 204, and supplied to the digital amplifier board 320 via the receiver 311. As a result, an acoustic beam having a designated directivity and direction is generated by the cylindrical speaker array unit 300.

  Next, signal processing performed in the DSP 201 to change the directivity and direction of the acoustic beam with time will be described. As described above, when the information indicating the directivity and direction of a certain acoustic beam changes with the passage of time, the CPU 101 newly instructs the directivity every time the indicated directivity and direction change. Then, an acoustic beam configuration parameter corresponding to the direction is obtained and sent to the DSP 201. In order to cope with such switching of the acoustic beam configuration parameters, the DSP 201 more specifically performs, for example, a delay process 2014, a windowing process 2015A, and a mapping process 2016 shown in FIG. 12 by a method shown in FIGS. Is running. FIG. 13 shows only a part related to the windowing process 2015A as an example, but the parts related to other windowing processes also have the same processing contents.

  First, the DSP 201 selects one of the register arrays RAij (i = 1 to 12, j = 1 to 36) and the register array RBij (i = 1 to 12, j = 1 to 36). A switch SW is provided. Here, each signal stored in each register of the register array selected by the switch SW is used for driving the speaker unit SPij (i = 1 to 12, j = 1 to 36).

  For example, the delay time DAij (i = 1 to 12, j = ja to jb), the window function value WAAij (i = 1 to 12, j = ja to jb) and the speaker unit according to the acoustic beam configuration parameters currently given. It is assumed that SPij (i = 1 to 12, j = ja to jb) is designated. In this case, the DSP 201 generates a plurality of delayed digital audio signals having delay times DAij (i = 1 to 12, j = ja to jb) from the oversampled signal S1, and generates window function values WAAij ( i = 1 to 12, j = ja to jb). Then, the plurality of windowed delayed digital signals obtained by multiplication are, for example, registered in the register RAij (i = 1 to 12, j = ja to jb) in the register array RAij (i = 1 to 12, j = 1 to 36). ). While this operation is performed, the switch SW selects the register array RAij (i = 1 to 12, j = 1 to 36), and the signal stored therein is the speaker unit SPij (i = 1 to 12, j = 1 to 36) (RAij mapping in FIG. 14).

  Thereafter, a new acoustic beam configuration parameter is generated by the CPU 101, and the delay time DBij (i = 1 to 12, j = ja ′ to jb ′), the window function value WBAij (i = 1 to 12, j = ja ′ to jb ′) and speaker unit SPij (i = 1 to 12, j = ja ′ to jb ′) are designated.

  In this case, while continuing the signal processing using the old acoustic beam configuration parameter, the delay processing and windowing processing using the new acoustic beam configuration parameter are executed, and the resulting delayed digital audio signal that has been windowed is obtained. Write to the register RBij (i = 1 to 12, j = ja ′ to jb ′) in the other register array RBij (i = 1 to 12, j = 1 to 36). Then, when this writing is completed, the switch SW is switched to the register array RBij (i = 1 to 12, j = 1 to 36) side. As a result, signals stored in the register array RBij (i = 1 to 12, j = 1 to 36) are used for driving the speaker unit SPij (i = 1 to 12, j = 1 to 36) (in FIG. 14). RBij mapping).

Thereafter, the same applies to the case where a new acoustic beam configuration parameter is generated by the CPU 101. While continuing the signal processing using the register array RBij (i = 1 to 12, j = 1 to 36) currently in use, The result of signal processing based on the new acoustic beam configuration parameter is written to the register array RAij (i = 1 to 12, j = 1 to 36), and then the register array is switched.
By repeating the operation as described above, signal processing corresponding to the acoustic beam configuration parameter that changes with time is advanced, and an acoustic beam whose directivity and direction change with the passage of time is generated.

  The operation of the DSP 201 has been described above using the digital audio signal S1 as an example, but the same processing is performed for the other digital audio signals S2 to S5. The signal processing results of the digital audio signals S1 to S5 are added for each speaker unit SPij in the addition processing 2018, and used for driving the cylindrical speaker array unit 300. Here, the CPU 101 in the present embodiment generates an acoustic beam configuration parameter for each of the digital audio signals S1 to S5 and supplies it to the DSP 201, and the DSP 201 provides signal processing for the digital audio signals S1 to S5 for each. Performed using the determined acoustic beam configuration parameters. Therefore, according to the present embodiment, the five types of acoustic beams having different directivities and directions can be generated by the cylindrical speaker array unit 300. In addition, in this embodiment, the cylindrical speaker array unit 300 is configured as illustrated in FIGS. 2 and 3, so that the directivity and direction of the acoustic beam can be arbitrarily set within a 360-degree omnidirectional range. Further, the directivity and direction of the acoustic beam can be changed in the axial direction of the cylindrical speaker array unit 300.

Second Embodiment
In the first embodiment, some speaker units SP that are likely to be useful for generating the instructed acoustic beam are selected from the speaker units SP of the cylindrical speaker array unit 300, and these speaker units SP are selected. Only worked. In the loudspeaker array system according to the second embodiment of the present invention, the loudspeaker unit SP that is not selected for generating the acoustic beam is operated as a wraparound sound canceling means. The configuration of this speaker array system is basically the same as that of the first embodiment. The difference from the first embodiment is that the CPU 101 not only performs the generation process of the acoustic beam configuration parameter but also generates a sound that cancels out the sneak sound to the speaker unit SP that is not selected for the generation of the acoustic beam. This is the point of controlling the DSP 201 for this purpose.

  FIG. 15 shows an operation example of the speaker array system according to the second embodiment of the present invention. In this example, using the speaker unit SP within the range indicated by the arrow Y1, an acoustic beam that causes a sound source to be heard at the position P is radiated into the area AR. As described in the first embodiment, by adjusting the delay time of the audio signal supplied to each speaker unit SP and performing an appropriate windowing process, the emission of sound waves to the outside of the area AR can be suppressed to some extent. it can. However, since the sound output from the speaker unit SP propagates in a relatively wide range, the sound inevitably reaches the area BR outside the area AR. Therefore, in the present embodiment, the canceling sound that cancels out the wraparound sound in the area BR using the speaker unit SP in the range indicated by the arrow Y2 that is not selected as the speaker unit SP for generating the acoustic beam is the area BR. Radiates in. There are the following three methods for generating the canceling sound.

a. First Cancellation Sound Generation Method In this method, the output sound of a certain speaker unit SP (hereinafter referred to as a wraparound sound generation speaker unit) within the range of the arrow Y1 becomes a wraparound sound, and a certain position in the area BR (hereinafter referred to as a wraparound). When it is known that the sound reaches the sound arrival position Q), a canceling sound is generated as follows. That is, as shown in FIG. 16, the delay time of the delayed audio signal supplied to the speaker unit SP in the arrow Y2 is adjusted so that the canceling sound having the opposite phase to the wrapping sound arrives at the wrapping sound arrival position Q. The CPU 101 generates a canceling sound parameter necessary for obtaining such a delayed audio signal and sends it to the DSP 201.

b. Second Canceling Sound Generation Method In this method, when it is known that the output sound of a wraparound sound generation speaker unit within the range of the arrow Y1 becomes a wraparound sound and reaches a wraparound sound arrival position Q in the area BR The canceling sound is generated as follows. First, the transfer function G1 of the path until the audio signal reaches the position in the area BR through the wraparound sound generating speaker unit is obtained in advance. Next, a certain speaker unit SP within the range indicated by the arrow Y2 is selected as a canceling sound speaker unit, and a transfer function G2 from the canceling sound speaker unit to the wraparound sound arrival position Q is obtained in advance. When an acoustic beam is radiated into the area AR based on a certain audio signal, the audio signal is subjected to a filtering process corresponding to the transfer function G1, as shown in FIG. generating an audio signal corresponding to, after inverting the phase of the audio signal, performs an inverse function G2 ^-1 equivalent filtering of the transfer function G2, and supplies to the speaker unit for canceling sound. By performing such processing, it is possible to supply a canceling sound having a phase opposite to that of the wraparound sound to the arrival position of the wraparound sound and cancel the wraparound sound.

c. Third canceling sound generation method In this method, when it is known that the output sound of a wraparound sound generation speaker unit within the range of the arrow Y1 becomes a wraparound sound and reaches a wraparound sound arrival position Q in the area BR A sensor for detecting the wraparound sound is arranged at the wraparound sound arrival position Q. In addition, a certain speaker unit SP within the range indicated by the arrow Y2 is selected as a canceling sound speaker unit. Then, when the acoustic beam is radiated into the area AR based on a certain audio signal, as shown in FIG. 18, the audio level is reduced so that the level of the wraparound sound at the wraparound sound arrival position Q detected by the sensor becomes a minimum value. The signal is subjected to adaptive filter processing and supplied to the canceling sound speaker unit. By performing such processing, the level of the wraparound sound at the wraparound sound arrival position Q can be kept low.

  The canceling sound generation method described above is particularly effective for canceling low-frequency wraparound sounds. Accordingly, the canceling sound generation method may be implemented only for the low frequency components.

<Other embodiments>
In addition to the embodiments described above, for example, the following embodiments are also conceivable in the present invention.
(1) In each of the above embodiments, a cylindrical speaker array is used. However, the shape of the speaker array may be, for example, a cone, a polyhedron, or a sphere.
(2) As shown in FIG. 19, for example, the cylindrical speaker array unit 300 is divided into two parts, and the two DSPs 201 </ b> A and 201 </ b> B share the process of generating a delayed digital audio signal applied to the speaker unit SP in the divided area. It may be. By doing so, it is possible to reduce the burden on each DSP and increase the processing speed.

1 is a block diagram illustrating a configuration of a speaker array system according to a first embodiment of the present invention. It is a top view which shows the structure of the cylindrical speaker array part in the speaker array system. It is a side view which shows the structure of the same cylindrical speaker array part. It is a figure explaining the generation principle of the acoustic beam in the embodiment. It is a figure explaining the generation principle of the acoustic beam in the embodiment. It is a figure explaining the generation principle of the acoustic beam in the embodiment. It is a figure explaining the generation principle of the acoustic beam in the embodiment. It is a figure explaining the generation principle of the acoustic beam in the embodiment. It is a figure explaining the generation principle of the acoustic beam in the embodiment. It is a figure which shows the example of the system using multiple said speaker array systems. It is a figure explaining the processing content of CPU in the same embodiment. It is a figure explaining the processing content of DSP in the same embodiment. It is a figure explaining the processing content of DSP in the same embodiment. It is a figure explaining the processing content of DSP in the same embodiment. It is a figure which shows operation | movement of the speaker array system which is 2nd Embodiment of this invention. It is a figure which shows the 1st example of the cancellation sound generation method in the embodiment. It is a figure which shows the 1st example of the cancellation sound generation method in the embodiment. It is a figure which shows the 1st example of the cancellation sound generation method in the embodiment. It is a figure which shows other embodiment of this invention.

Explanation of symbols

101 ... CPU, 201 ... DSP, 300 ... cylindrical speaker array unit.

Claims (4)

A speaker array composed of a plurality of speaker units arranged to form a convex surface;
Based on the information specifying the directivity and direction of the acoustic beam output from the speaker array, a plurality of speaker units used for the output of the acoustic beam in the speaker array are selected, and a plurality of delayed audio signals supplied to these speaker units are selected. A computing means for computing the delay time;
Signal processing means for performing signal processing for obtaining the plurality of delayed audio signals from a common audio signal.   When information specifying the directivity and direction of a plurality of types of acoustic beams generated from a plurality of types of audio signals is given, the calculation means calculates the delay times of the plurality of delayed audio signals for each acoustic beam. The signal processing means performs signal processing for obtaining a plurality of delayed audio signals for outputting each of the acoustic beams from the speaker array from each of the plurality of audio signals. Item 2. The speaker array system according to Item 1. The signal processing means includes
Band dividing means for dividing the audio signal into audio signals of a plurality of frequency bands;
Delay means for generating a plurality of delayed audio signals from each of a plurality of bands of audio signals obtained by the band dividing means;
The speaker array system according to claim 1, further comprising: a windowing unit that multiplies the plurality of audio signals of the plurality of bands by a window function corresponding to each band.   The apparatus further comprises canceling sound generating means for generating a canceling sound that cancels a wraparound sound generated outside the acoustic beam using a speaker unit that is not selected for use in outputting the acoustic beam in the speaker array. The speaker array system according to any one of claims 1 to 3.

Images