Patent Translate Powered by EPO and Google Notice This translation is machine-generated. It cannot be guaranteed that it is intelligible, accurate, complete, reliable or fit for specific purposes. Critical decisions, such as commercially relevant or financial decisions, should not be based on machine-translation output. DESCRIPTION JP2009010491 A speaker array device, a microphone array device, and a signal processing method capable of performing directivity control in a wide range of frequency bands by a process with low load. A speaker array device (1) of the present invention divides an audio signal (Sin) input from a signal input unit (5) into a plurality of frequency bands in a signal dividing unit (4). By emitting an audio signal that has been subjected to amplification processing based on the window function set in each of the amplification units of the signal processing unit 3, desired directivity characteristics can be obtained in a wide frequency band. . Further, since it is not necessary to finely divide the frequency band, the amount of calculation of signal processing in the signal processing unit 3 can be reduced, and directivity control in a wide frequency band can be performed by a process with a small load. [Selected figure] Figure 1 Speaker array device, microphone array device, and signal processing method [0001] The present invention relates to a technique for improving the directivity of a speaker array. [0002] As a speaker system capable of improving directivity, that is, a speaker system having so-called narrow directivity, for example, there is a speaker array. The speaker array can control the directivity of the sound by controlling the amplitude, phase, 09-05-2019 1 and the like of the sound emitted from the individual speakers, and can emit the beamed sound to a desired location. By beaming the sound, it is possible to transmit the sound with little attenuation of the volume even at a distant place, so it is frequently used in a large hole or the like. [0003] On the other hand, in the speaker array, the frequency band of sound tends to be narrow due to the control of the directivity state. Therefore, in order to control the pointing state over a wide frequency band, a technique for optimizing and controlling the phase and amplitude of the input audio signal at each frequency point using FIR (Finite Impulse Response) is disclosed. (For example, Patent Document 1, Non-Patent Document 1, and Non-Patent Document 2). Patent Document 1: Japanese Patent Application Laid-Open No. 2-239798 Tetsuguchi Taniguchi, Kiyoshi Nishikawa, Masaki Amano, "Wide-band beam forming method by multiple band division using a Dolph-Chebyshev spatial filter", Journal of the Institute of Electronics, Information and Communication Engineers 1995/12 Vol. J78-A No. 12 p1576-p1584 Mitsuya Ohya, Kiyoshi Nishikawa, “Any Beam Directional Directional Array Speaker with Band Division Design”, 10th Digital Signal Processing Symposium 1995 / 11-2 p59-p64 [0004] However, when the phase and amplitude are controlled at individual frequency points using an FIR filter, the amount of calculation becomes enormous and the processing load on a DSP (Digital Signal Processor) or the like becomes extremely large. [0005] The present invention has been made in view of the above-described circumstances, and provides a speaker array device, a microphone array device, and a signal processing method that can perform directivity control in a wide range of frequency bands with a process with low load. With the goal. [0006] In order to solve the problems described above, the present invention divides a plurality of divided audio signals having different frequency bands by dividing a plurality of sound emitting means and an input audio signal according to one or more preset frequency characteristics. And a signal processing unit having a plurality of amplifying unit groups each having a plurality of amplifying units each set with an amplification factor based on a preset window function, each of 09-05-2019 2 the amplifying unit groups comprising: Among the plurality of divided audio signals generated by the signal dividing means, divided audio signals of a predetermined frequency band and of a frequency band different from that of the other amplification means group are supplied, and each of the above-mentioned respective amplification means groups The amplification means is configured to release the signal obtained by amplifying the divided audio signal supplied to the amplification means group by the set amplification factor, connected to the amplification means. The plurality of sound emitting means emit sound beams having predetermined directivity characteristics by emitting the signals supplied from the respective amplifying means, and the plurality of sound emitting means are preset to the respective amplifying means groups. The window function is such that the number of intensity minima in the stopband of the directivity characteristic increases in order from the amplification means group which is a frequency band in which the frequency band of the divided audio signal supplied is low among the plurality of amplification means groups. The speaker array device is characterized in that one or more frequency characteristics of the signal dividing means are set such that the intensity of a predetermined angle does not exceed a predetermined value in the directivity characteristics. . [0007] In another preferred embodiment, the apparatus further comprises control means for changing the directivity characteristic, and each of the sound emitting means performs delay processing on the signal supplied from each of the amplification means, and delay processing. The control unit may change the directivity characteristic by controlling the delay amount of the delay means constituting each of the sound emitting means. [0008] In another preferred embodiment, the apparatus further comprises control means for changing the directivity characteristics, each amplification means group further comprises delay means connected to each of the amplification means to perform delay processing, Each of the delay means constituting each amplification means group supplies a signal obtained by delaying the divided audio signal supplied to the amplification means group to amplification means connected to the delay means which has performed the delay processing. The amplification means constituting the amplification means group are replaced with the divided audio signals supplied to the amplification means group, and the signals supplied from the delaying means are set at the amplification factor set. The amplified signal may be supplied to the connected sound output means, and the control unit may change the directivity by controlling the delay amount of the delay means. [0009] Further, in another preferable aspect, when the directivity characteristic is changed by the 09-05-2019 3 control unit controlling each of the delay units, one or more frequency characteristics set in the signal division unit are changed. The means may be further provided. [0010] Further, according to the present invention, there are provided a plurality of signal processing means including a plurality of sound collecting means, and a plurality of amplification means groups each having a plurality of amplification means each having an amplification factor set based on a preset window function. A predetermined frequency band is set among a plurality of frequency bands obtained by dividing in the above frequency characteristics, and a plurality of band selection means for attenuating and outputting amplitudes of frequency bands other than the frequency band of the supplied signal And signal dividing / synthesizing means having an adding means for adding the signals output from the plurality of band selecting means, each of the sound collecting means generates an audio signal based on the collected sound, and The amplification means is supplied with an audio signal generated by the sound collection means connected to the respective amplification means, and each amplification means group is adapted to the supplied audio signal. The signals amplified at the set amplification factor are added by the respective amplification means constituting the width means group, and the added signal is supplied to the band selection means connected to the amplification means group, and the plurality of the plurality of The predetermined frequency bands set in the band selection means are respectively different frequency bands, and the sound collection performed by the plurality of sound collection means has predetermined directivity characteristics, and the windows preset in each of the amplification means The function is such that the number of intensity minima in the stopband of the directivity characteristic increases in order from the amplification means group connected to the band selection means which is a low frequency band and the frequency band set among the plurality of band selection means The microphone is characterized in that one or more frequency characteristics of the signal division and synthesis means are set such that the intensity of a predetermined angle does not exceed a predetermined value in the directivity characteristic. To provide a ray equipment. [0011] Further, the present invention is a signal processing method used in a speaker array device having a plurality of sound emitting means, which divides an audio signal to be input into one or more preset frequency characteristics, thereby to make the frequency band different. And a signal processing step of generating a plurality of divided audio signals, and a plurality of amplification step groups each including a plurality of amplification steps each having an amplification factor set based on a preset window function. The process group is a predetermined frequency band among a plurality of divided audio signals generated in the signal division process, and a divided audio signal of a frequency band different from other amplification process groups is supplied, and each of the amplification process groups In each of 09-05-2019 4 the amplification processes, the divided audio signal supplied to the amplification process group is amplified by the set amplification factor. The plurality of sound emitting means are supplied to sound emitting means corresponding to the amplification process, and the plurality of sound emitting means emit an acoustic beam having a predetermined directivity characteristic by emitting the signals supplied in the respective amplification processes, The window function preset in each amplification process group is the intensity in the stopband of the directivity characteristic in order from the amplification process group in which the frequency band of the divided audio signal supplied among the plurality of amplification process groups is low. The number of local minima is set to increase, and in the directivity characteristic, one or more frequency characteristics in the signal division process are set such that the intensity at a predetermined angle does not exceed a predetermined value. To provide a signal processing method. [0012] Further, the present invention is a signal processing method for use in a microphone array apparatus having a plurality of sound collection means, which comprises a plurality of amplification processes each including a plurality of amplification processes each having an amplification factor set based on a preset window function. A predetermined frequency band is set among a plurality of frequency bands obtained by dividing a plurality of signal processing steps and one or more preset frequency characteristics, and a frequency band other than the frequency band of the supplied signal is set. A plurality of band selection processes for attenuating and outputting the amplitude, and a signal division and synthesis process including an addition process for adding the signals output from the plurality of band selection processes; An audio signal is generated on the basis of the respective amplification processes, and the respective amplification processes are supplied with audio signals generated by the sound collection means corresponding to the respective amplification processes, The group adds the signal amplified by the amplification factor set by the amplification process in the amplification process group to the supplied audio signal, and adds the signal obtained by the addition to the band corresponding to the amplification process group The predetermined frequency bands supplied to the selection process and set in the plurality of band selection processes are respectively different frequency bands, and the sound collection performed by the plurality of sound collection means has a predetermined directivity characteristic, The window function set in advance for each amplification process is, in order from the amplification process group corresponding to the band selection process having a low frequency band among the plurality of band selection processes, in the stopband of the directivity characteristic. The number of local minima of the intensity is set to increase, and in the directivity characteristic, one or more frequencies in the signal division and combining process such that the intensity at a predetermined angle does not exceed a predetermined value. To provide a signal processing method characterized by sex is set. 09-05-2019 5 [0013] According to the present invention, it is possible to provide a speaker array device, a microphone array device, and a signal processing method capable of performing directivity control in a wide range of frequency bands by a process with low load. [0014] Hereinafter, an embodiment of the present invention will be described. [0015] Embodiment First, the configuration of the speaker array device 1 according to the present embodiment will be described. FIG. 1 is a block diagram showing the configuration of the speaker array device 1. The sound output unit 2 includes non-directional speakers 2-1, 2-2,..., 2-n arranged linearly facing the same direction, and the audio supplied from the signal processing unit 3 Sound the signal. The sound emitting unit 2 emits an audio signal subjected to signal processing as described later from each of the speakers 2-1, 2-2,. The beam can be emitted. [0016] The signal processing unit 3 includes amplification units 31, 32, ..., 35 and an adder 30, as shown in FIG. Here, the configuration of each of the amplification units 31, 32, ..., 35 will be described using the amplification unit 31 as an example. 09-05-2019 6 The amplification unit 31 has amplification circuits 31-1, 31-2, ..., 31-n as shown in FIG. Each of the amplifier circuits 31-1, 31-2,..., 31-n amplifies the audio signal Sa-1 input from the signal division unit 4 at the set amplification factor, and the amplifier circuit 31-1 The signal is supplied to the speaker 2-1, and the amplification unit 31-2 is supplied to the speakers 2-1, 22,..., 2-n respectively connected to be supplied to the speaker 2-2. Hereinafter, the aspect of the amplification factor set to each amplifier circuit 31-1, 31-2, ..., 31-n is called a window function. In this manner, each of the amplification units 31, 32,..., 35 transmits the five systems of audio signals Sa-1, Sa-2,..., Sa-5 supplied from the signal division unit 4. The amplification processing is performed based on the window functions set respectively, and is supplied to each of the speakers 2-1, 2-2,. Then, the audio signals output to the same speaker are added by each adder 30 and supplied to the speaker. Therefore, each of the speakers 2-1, 2-2,..., 2-n emits an audio signal obtained by adding the audio signal output from each of the amplification units 31, 32,. It will be. The window function is set by the control of the control unit 6, but the details will be described later. [0017] Returning to FIG. 1, the description will be continued. The signal division unit 4 divides the audio signal Sin input from the signal input unit 5 into a plurality of frequency bands to generate audio signals Sa-1, Sa-2, ..., Sa-5, and a signal processing unit , And 35, respectively. Here, the configuration of the signal division unit 4 will be described with reference to FIG. As shown in FIG. 4, the signal dividing unit 4 includes an LPF (Low Pass Filter, low pass filter) 4-1, a BPF (Band Pass Filter, band pass filter) 4-2, 4-3, 4-4, an HPF High Pass Filter, high pass filter) 4-5. The LPF 4-1 attenuates the amplitude of the set frequency f1 or more with respect to the input 09-05-2019 7 audio signal Sin, and supplies the audio signal Sa-1 to the amplification unit 31 of the signal processing unit 3 as the audio signal Sa-1. The BPFs 4-2, 4-3, and 4-4 set frequency bands for the input audio signal Sin (for the BPF 4-2, for the frequency band from the lower limit frequency f1 to the upper limit frequency f2, BPF 4-3) Attenuates amplitudes other than the frequency band from the lower limit frequency f2 to the upper limit frequency f3 and the frequency band from the lower limit frequency f3 to the upper limit frequency f4 in the BPF 4-4 to attenuate as audio signals Sa-2, Sa-3 and Sa-4 The signal is supplied to the amplification units 32, 33, 34 of the signal processing unit 3. The HPF 4-5 attenuates the amplitude of the set frequency f4 or less with respect to the input audio signal Sin, and supplies it to the amplification unit 35 of the signal processing unit 3 as an audio signal Sa-5. Here, the relationship between the set frequencies is f1 <f2 <f3 <f4, and if the audio signals supplied from the signal division unit are arranged in order from the lower frequency band, the audio signals Sa-1 and Sa-2 are obtained. , ..., Sa-5 in order. Although the frequency is set by the control of the control unit 6, the details will be described later. [0018] Returning to FIG. 1, the description will be continued. The control unit 6 controls each unit of the speaker array device 1 as described above. This control may be performed based on the input setting value when the user operates the operation unit 7 or may be performed based on the setting value stored in the storage unit 8. Here, the setting value is, for example, a window function set in each of the amplification units 31, 32,..., 35 of the signal processing unit 3, and frequencies f1, f2, f3 and f4 set in the signal division unit 4. , Directivity characteristics of an acoustic beam, and the like. Note that the storage unit 8 combines these setting values and stores a plurality of setting value sets as a table, so that the user operates the operation unit 7 and the setting values stored in the storage unit 8 are stored. By selecting one set of set values from among sets, the control unit 6 can also control each unit based on the set values of the selected set. [0019] Next, the window function set in each of the amplification units 31, 32, ..., 35 of the signal processing unit 3 under the control of the control unit 6 will be described. First, window functions to be candidates set for each of the amplification units 31, 32,..., 35 are determined. A plurality of candidate window functions are known using nonlinear optimization or the least squares method at each frequency (preferably several Hz unit is desired) under predetermined conditions in the speaker array device 1 such as the distance between speakers Filter design 09-05-2019 8 method (hereinafter, non-linear optimization will be described as an example, but it may be a well-known filter design method using the least squares method or the like). For example, the directivity characteristic of the acoustic beam in the case where the sound of 879 Hz is emitted from the sound emitting unit 2 using the window function obtained by non-linear optimization at 879 Hz is the directivity characteristic as shown in FIG. . Here, FIG. 5 is a polar coordinate graph in which the radial direction represents intensity (in dB when referred to the 0 degree direction) and the angular direction is 0 degrees in the front direction of the speaker as shown in FIG. Similarly, the directivity characteristics of the acoustic beam when the sound at 872 Hz is emitted from the sound emitting unit 2 using the window function obtained by non-linear optimization at 872 Hz is as shown in FIG. 5B. It becomes a characteristic. [0020] Thus, at near frequencies, the number of local minima at which the intensity of the stopband (an angular region other than the main lobe having a large intensity in the 0 degree direction) in the directivity characteristics of the acoustic beam becomes minimal (in FIG. If the number of local minimum points from 4 degrees to 90 degrees is 4 or less, and simply referred to as the number of local minimum points, the window function which formally means the number of local minimum points from 0 degree to 90 degrees is the same can get. Among the window functions obtained for each frequency in this manner, the window function obtained by the non-linear optimization is set to one amplification unit at a frequency at which the number of local minima in the directional characteristic becomes the same. Classify as a candidate. Then, among window function candidates set in one amplification unit, a target directivity characteristic, for example, a window function having a main lobe width close to the target width is selected from the window functions as candidates, and this is selected. It is determined as a window function to be set in the amplification unit. For example, when comparing the directivity characteristics of an acoustic beam at 879 Hz and the directivity characteristics of an acoustic beam at 872 Hz, it is a window function that can realize a state in which the main lobe width is narrow while the frequency at 872 Hz is lower. The window function determined at 872 Hz is determined as the window function set in the amplification unit. Then, with respect to other window function candidates related to directional characteristics with different numbers of local minimum points, the window functions to be set to one amplification unit are determined in the same manner. By determining the window function in this manner, the main lobe width can be made as narrow as possible because the window function that most satisfies the condition can be used if the number of local minima is the same. In addition, since the same window function can be used for the frequencies having the same minimum number, the number of band divisions can be reduced. [0021] 09-05-2019 9 As shown in FIG. 7, FIG. 8 and FIG. 9, the directivity characteristics of the acoustic beam in the case where the sound of the frequency for which the window function is determined is emitted from the sound emitting unit 2 using the window function thus determined. It uses a window function whose frequency is determined at 126 Hz, 327 Hz, and 585 Hz, respectively. Here, FIG. 5 shows that the number of minimum points is 4 but FIG. 7 shows that the number of minimum points is 1, FIG. 8 shows that the number of minimum points is 2 and FIG. 9 shows that the number of minimum points is 3 The window function determined is used. [0022] Then, the frequency band of the input audio signal is set by the control unit 6 from the low amplification unit in order from the window function relating to the directivity characteristic with the smaller number of minimum points. For example, the window function obtained at 126 Hz (the number of minimum points is 1) is set in the amplification unit 31, and the window function obtained at 327 Hz (the number of minimum points is 2) is set in the amplification unit 32. In this manner, the window function is set in each of the amplification units 31, 32, ..., 35 by the control of the control unit 6. Each of the speakers 2-1, 2-2,..., 2-n in the present embodiment is a nondirectional speaker, but may be a speaker having a directional characteristic. In this case, the window function may be determined according to the directivity characteristics of each of the speakers 2-1, 2-2,. [0023] Next, the frequencies f1, f2, f3, and f4 set in the signal dividing unit 4 by the control of the control unit 6 will be described. The audio signal Sa-1 whose frequency band upper limit is determined by the frequency f1 is supplied to the amplification unit 31. The window function set in the amplifying unit 31 is a window function determined at 126 Hz, and when the sound of 126 Hz is emitted from the sound emitting unit 2, as a sound beam of directional characteristics as shown in FIG. 7 as described above Released. When sounds of various frequencies are emitted using this window function, the width of the main lobe widens at frequencies lower than 126 Hz and the width of the main lobe narrows at high frequencies. On the other hand, on the high frequency band side, disturbance of directivity characteristics such as an increase in intensity of side lobes other than the main lobe occurs. For example, when a 2000 Hz sound is emitted from the sound emitting unit 2 using a window function determined at 872 Hz, an acoustic beam having a disordered directivity as shown in FIG. 10 is emitted. 09-05-2019 10 [0024] As described above, when the intensity of the side lobes is increased, the directivity characteristic of the acoustic beam is disturbed. Therefore, when using the window function set in each of the amplification units 31, 32,. It is necessary to adjust the frequency characteristics at 31, 32, ..., 35. Although there are various methods for this method, the frequency characteristic is determined in the range in which the intensity in a predetermined angular direction does not exceed a predetermined value. In the present embodiment, for example, the predetermined angle is 90 degrees, and the predetermined value is -10 dB. The predetermined angle and the predetermined value do not have to be this value, and may be appropriately changed to obtain a desired directional characteristic. The change may be made by the user operating the operation unit 7. [0025] Specifically, the frequency f1 is determined as follows. The directivity characteristics of the acoustic beam when the sound of 130 Hz and 140 Hz is emitted from the sound emission unit 2 using the window function set in the amplification unit 31 are the directivity characteristics as shown in FIG. 11 and FIG. 12. As described above, it can be seen that the width of the main lobe is narrower at 130 Hz than at 126 Hz, and the width of the main lobe is narrower at 140 Hz. On the other hand, it can be seen that the intensity of the side lobes increases as the frequency increases. Since the intensity at 90 degrees is -14 dB in FIG. 11 and -10 dB in FIG. 12, the input audio signal Sa-1 is obtained when the window function set in the amplification unit 31 is used. The upper limit of the frequency band of is determined as 140 Hz. Thus, by the control of the control unit 6, the frequency f1 in the signal division unit 4 is set to f1 = 140 Hz. In this case, at the frequency f1, the output of the window function (corresponding to the amplification unit 31) is slightly attenuated but includes the contribution of the output of the window function (corresponding to the amplification unit 32) in the upper frequency band. There is. Therefore, more strictly, in the frequency characteristic of the audio signal output from the signal dividing unit 4 and the result of processing with each window function, the intensity in a specific angular direction is taken into consideration, taking into consideration the directivity characteristic of the speaker if necessary. The frequency characteristic of each filter in the signal dividing unit 4 may be determined at the limit of the range in which the value does not exceed the predetermined value. Here, when designing as a linear phase FIR filter using a method such as a window function method or Fourier series approximation, the disturbance of the characteristic in the transient region in the frequency characteristic of the audio signal divided by the signal division unit 4 is reduced. Can. 09-05-2019 11 [0026] The frequency f2 related to the frequency characteristic of the audio signal Sa-2 is determined as described above using the window function set in the amplification unit 32, and is set in the signal division unit 4 by the control of the control unit 6. Ru. The frequencies f3 and f4 are similarly set. Thus, under the control of the control unit 6, the signal division unit 4 sets the frequencies f1, f2, f3 and f4. [0027] Next, for the operation of the speaker array device 1 in which the window function is set in the signal processing unit 3 and the frequencies f1, f2, f3 and f4 are set in the signal division unit 4 as described above, an audio signal from the signal input unit 5 A description will be given until Sin is input and sound is emitted from the sound emission unit 2. [0028] The audio signal Sin input from the signal input unit 5 is output to the signal division unit 4. The signal dividing unit 4 distributes the audio signal Sin to the LPF 4-1, the BPF 4-2, 4-3, 4-4, and the HPF 4-5, which are different based on the set frequencies f1, f2, f3, and f4. The audio signals Sa-1, Sa-2, ..., Sa-5 divided into frequency bands are generated and supplied to the signal processing unit 3. [0029] The audio signal Sa-1 supplied from the signal division unit 4 is supplied to the amplification unit 31 of the signal processing unit 3. The audio signal Sa-1 supplied to the amplification unit 31 is amplified at each of the amplification circuits 31-1, 31-2, ..., 31-n based on the set window function, and each speaker 2-1 , 2-2,..., 2-n. Similarly, the other audio signals Sa-2, Sa-3, Sa-4, Sa5 are amplified based on the window function set in the amplification units 32, 33, 34, 35, respectively, and each speaker 2 -1, 2-2, ..., 2-n. At this time, the audio signals supplied from the respective amplification units 31, 32, ..., 35 to the same speaker are respectively added by the 09-05-2019 12 adder 30 and emitted from the speaker. [0030] As described above, the audio signal Sin input from the signal input unit 5 is divided into a plurality of frequency bands in the signal division unit 4, and the respective amplification units 31 of the signal processing unit 3 with respect to audio signals in different frequency bands. By emitting an audio signal that has been subjected to amplification processing based on the window function set to 32, ..., 35, desired directivity characteristics can be obtained in a wide frequency band. . Further, since it is not necessary to finely divide the frequency band, the amount of calculation of signal processing in the signal processing unit 3 can be reduced, and directivity control in a wide frequency band can be performed by a process with a small load. [0031] Although the embodiments of the present invention have been described above, the present invention can be implemented in various aspects as follows. [0032] <Modification 1> In the embodiment, the signal dividing unit 4 divides the audio signal Sin by the frequencies f1, f2, f3, and f4 of 4 so that the audio signals Sa-1 and Sa-2 in the frequency band of 5 can be obtained. ,..., Sa-5, but the number of audio signals generated by division is not limited to five, and may be any number as long as it is two or more. In this case, the number of amplification units of the signal processing unit 3 may be increased or decreased according to the number of divided audio signals, and the number of divisions of the audio signal Sin and the number of amplification units are realized by the control of the control unit 6 Just do it. This may be determined by calculating the required number of divisions based on the desired directional characteristics, the number of speakers of the sound emitting unit 2, the result of the non-linear optimization, and the like. [0033] 09-05-2019 13 <Modification 2> In the embodiment, the directivity control is performed by amplification processing based on the window function set in each of the amplification units 31, 32, ..., 35 of the signal processing unit 3. Although it was not possible to control the direction of the lobe (hereinafter referred to as the pointing direction), it is also possible to control the pointing direction. In this case, as shown in FIG. 13, a delay unit 21 having delay circuits 21-1, 21-2,..., 21n is provided in the sound emitting unit 2, and each delay circuit 21-1, 21-2. ,..., 21-n may perform delay processing of the delay amount set for the audio signal supplied from the signal processing unit 3. Here, the control unit 6 calculates the delay amount based on the pointing direction instructed by the user operating the operation unit 7, and the delay unit 21-1 of the delay unit 21 is controlled by the control unit 6. , 21-2, ..., 21-n. The pointing direction does not have to be a user's instruction, and information such as a change in pointing direction may be stored in the storage unit 8 and the pointing direction may be set according to the information. [0034] In addition, when the pointing direction of the 126 Hz acoustic beam as shown in FIG. 7 is changed by −5 degrees, for example, the intensity at 90 degrees may largely change as shown in FIG. 14 by changing the pointing direction. is there. In this case, the frequencies f1, f2, f3, and f4 set in the signal dividing unit 4 may be changed so as to satisfy the conditions described above, and in the case of changing them, the control of the control unit 6 The frequencies f1, f2, f3 and f4 set by the above may be changed. The window function set in each of the amplification units 31, 32, ..., 35 of the signal processing unit 3 may be set again by the method described above. Also in this case, the window function set by the control of the control unit 6 may be changed. [0035] <Modification 3> In the case where the delay unit 21 is provided as in the modification 2, the delay circuits 21-1, 21-2, ..., 21-n are delayed so as to form the focal point of the acoustic beam. By setting the amount, control of the directivity in the embodiment can also be performed, but in that case, disturbance of the directivity of the main lobe may occur in a high frequency band. In such a case, the signal input unit 5 may be frequency-dependent to have the function of an equalizer capable of adjusting the amplitude, and the frequency dependence of the amplitude may be adjusted so that the disturbance of the directivity characteristic does not occur. The function of the equalizer is provided to the LPF 4-1, BPF 4-2, BPF 4-2, 4-3, 4-4, and HPF 4-5 of the signal dividing unit 4, and the amplitude is adjusted in a predetermined frequency band not to be attenuated. May be 09-05-2019 14 [0036] <Modification 4> In the embodiment, the signal dividing unit 4 uses the LPF 4-1 to obtain the audio signal Sa-1 of the lowest frequency band, and obtains the audio signal Sa-5 of the highest frequency band. Although HPF 4-5 was used for this purpose, one or both of them may be BPF. In this case, the lower limit of the frequency band needs to be determined for the BPF used instead of the LPF 4-1, and the upper limit of the frequency band needs to be determined for the BPF used instead of the HPF 4-5. The upper and lower limits may be determined based on the audible range of sound, may be determined from the used frequency band based on the frequency characteristic of the audio signal Sin, and may be a desired value. This value may be determined in advance, or the user may designate a value by operating the operation unit 7. Further, in the case of determining based on the frequency characteristics of the audio signal Sin, the signal input unit 5 is provided with measurement means for measuring the frequency characteristics of Sin of the audio signal, and the control unit 6 measures the frequency characteristics measured by the measurement means It may be determined based on [0037] <Modification 5> In the embodiment, the window function set in each of the amplification units 31, 32, ..., 35 (in particular, the amplification unit in a high frequency band, and the amplification unit 35 in the embodiment) The width of the main lobe of the acoustic beam may be too narrow. In this case, as shown in FIG. 15, delay circuits 350-1, 350-2,..., 350-n for delaying the audio signal input to the amplification unit 35 are provided to adjust directivity characteristics. You may do so. In this way, delay processing can be performed for each frequency band. Note that such a delay circuit may be provided not only in an amplification unit that amplifies an audio signal in a high frequency band but also in a plurality of amplification units. Also, phase interference occurs in the overlapping frequency band between the sound emitted and delayed by the delay circuits 350-1, 350-2,..., 350-n and the sound emitted without the delay processing. If this happens, as shown in FIG. 16, delay circuits 310, 320, 330, 340 for delaying the input audio signal are provided to adjust the delay amount in each delay circuit 310, 320, 330, 340. Thus, the influence of phase interference may be reduced. If the delay circuits as described above are provided in all the amplification units, it is possible to change the directivity direction of the acoustic beam as shown in the second modification. When the directivity characteristic is disturbed, as described in the third modification, the signal input unit 5 or the like may have the function of an equalizer to adjust the frequency dependency of the amplitude. 09-05-2019 15 [0038] <Modification 6> In the embodiment, each of the amplification units 31, 32, ..., 35 has the same number of amplification circuits as the number of speakers included in the sound emission unit 2, but the windows set When the function has symmetry, the number of amplification circuits can be reduced according to the symmetry, and the amount of calculation of processing in the signal processing unit 3 can be reduced. For example, when the number of speakers is 11, and the output is symmetrical with respect to the speakers 2-6, as shown in FIG. 17, the amplification unit 31 includes amplification circuits 31-1, 31-2,. ..., to have a 31-6, respectively may be to also output from the amplifier circuits with respect to the speaker to be symmetrical output, in this way, the number of the amplifier circuit is reduced from 11 to 5 The processing load of the signal processing unit 3 can be reduced. [0039] <Modification 7> In the embodiment, although the speakers 2-1, 2-2,..., 2-n of the sound emitting unit 2 used the same speaker, each speaker 2-- of the sound emitting unit 2 1, 2-2,..., 2-n, for example, small diameter speakers 2-1, 2-2,. A speaker different from, for example, the largediameter speakers 2-m + 1, 2-m + 2,. In this case, as shown in FIG. 18, for example, the audio signals Sa-1 and Sa-2 of a relatively low frequency band are supplied to the amplification units 31 and 32 of the signal processing unit 3, so the speaker 2 The amplified audio signal is supplied to −m + 1, 2-m + 2,..., 2-n, and audio signals Sa-3, Sa of relatively high frequency bands are supplied to the amplification units 33, 34, 35. Since -4 and Sa-5 are supplied, the amplified audio signal may be supplied to the speakers 2-1, 2-2,..., 2-m. In this way, the width of directional control of the acoustic beam can be expanded, and for example, the width of the main lobe can be further narrowed even for an acoustic beam in a low frequency band. [0040] <Modification 8> In the embodiment, although amplification processing is performed on audio signals of all frequency bands in the signal processing unit 3, amplification processing is performed only on audio signals of some frequency bands. You may do so. In this case, for audio signals in the frequency band for which amplification processing is not performed in the signal processing unit 3, signal processing is performed to change the phase and amplitude with the FIR filter, thereby performing directivity control of the acoustic beam in the frequency band. May be For example, when the signal processing is performed only on the audio signal Sa-5 by the FIR 09-05-2019 16 filter, the signal processing is further performed with frequency dependency and the height of the audio signal output to the speaker located at the end is high. Frequency components can also be attenuated. As described above, by performing signal processing with the FIR filter, it is possible to change the frequency characteristics of the audio signal to be output for each speaker, so it is possible to obtain more desired directional characteristics. [0041] <Modification 9> In the embodiment, the window function set in each of the amplification units 31, 32, ..., 35 of the signal processing unit 3 is a candidate classified according to the number of local minimum points of directivity characteristics. The window function is determined for each classification, and the frequency band of the supplied audio signal is set by the control unit 6 in order from the window function relating to the directional characteristic with the smaller number of local minima. At this time, the number of local minimum points is assumed to increase by one since the window functions relating to the directivity characteristic having a small number of local minimums are sequentially set, but may not necessarily increase by one. For example, the window function set in the amplification unit 32 may not be the window function determined at 327 Hz (the number of minimum points is 2), but may be the window function determined at 585 Hz (the number of minimum points is 3). As a result, the frequency band of the audio signal Sa-2 becomes a frequency band (frequency f1 to f3) obtained by combining the audio signals Sa2 and Sa-3 in the embodiment, and the audio signal has a wide frequency band. In the window function related to 585 Hz, since the upper limit of the frequency band is originally the frequency f3, there is no influence of the disturbance of the directional characteristics. In this way, the frequency band handled by one amplification unit becomes wide, so the number of amplification units can be reduced, and the processing load of the signal processing unit 3 can be reduced. [0042] <Modification 10> In the embodiment, the control unit 6 controls the signal processing unit 3 based on the setting value input by the user operating the operation unit 7 and the setting value stored in the storage unit 8. The division unit 4 is controlled to set the window function to the signal processing unit 3 and the frequencies f1, f2, f3 and f4 to the signal division unit 4, but as described above based on the desired directivity characteristic The window function and the frequencies f1, f2, f3 and f4 may be calculated and set as In this case, the desired directional characteristic may be input by the user operating the operation unit 7. As described above, if the window function and the frequency for dividing the audio signal are calculated by the method as 09-05-2019 17 described in the embodiment, the value calculated in advance may be set as the setting value, and the control unit 6 may The calculated value may be used as the setting value. [0043] <Modification 11> In the embodiment, although the speaker array device 1 that emits an acoustic beam having a desired directional characteristic is used, the microphone array device 100 may be a directional microphone having a desired directional characteristic. In this case, the configuration as shown in FIG. 19, FIG. 20, FIG. 21, and FIG. 22 may be adopted. Hereinafter, the microphone array device 100 will be described. [0044] As shown in FIG. 19, the sound collection unit 9 having omnidirectional microphones 9-1, 9-2,..., 9-n includes the microphones 9-1, 9-2,. An audio signal relating to the sound picked up by -n is generated and supplied to the signal processing unit 13. The signal processing unit 13 has amplification units 131, 132,..., 135 as shown in FIG. The amplification unit 131 includes amplification circuits 131-1, 131-2,..., 131-n and an adder 1310 as shown in FIG. Each of the amplifier circuits 131-1, 131-2, ... 131-n is a microphone 9-1, 9-2, ..., with an amplification factor based on the window function set as described in the embodiment. The audio signal supplied from 9-n is amplified and output. Then, the audio signal output from each of the amplifier circuits 131-1, 131-2,... 131-n is added in the adder 1310, and is added to the signal division combining unit 14 as the audio signal Sb-1. [0045] The signal division combining unit 14 includes an LPF 14-1, BPFs 14-2, 14-3, 14-4, an HPF 14-5, and an adder 140, as shown in FIG. The LPF 14-1, BPF 14-2, 14-3, 14-4, and HPF 14-5 are set in the same manner as the LPF 4-1, BPF 4-2, 4-3, 4-4, and HPF 4-5 described in the embodiment. Signal processing based on the frequencies f1, f2, f3 and f4 is performed on the supplied audio signal. Here, the audio signal Sb-1 output from the signal processing unit 13 is supplied to the LPF 14-1, the audio signal Sb-2 to the BPF 14-2, and the audio signal Sb-3 to the BPF 14-3. -4 is supplied to the BPF 14-4, and the audio signal Sb-5 is supplied to the HPF 14-5. Then, by being subjected to signal processing by the LPF 14-1, BPF 14-2, 14-3, 14-4, and HPF 14-5, audio signals having different frequency bands are added by the adder 140 and output as an audio 09-05-2019 18 signal Sout. It is output from the unit 15. [0046] Hereinafter, the window functions set in the respective amplification units 131, 132,..., 135 of the signal processing unit 13 and the frequencies f1, f2, f3, and f4 set in the signal division combining unit 14 will be described in order. First, the window function set in each of the amplification units 131, 132,..., 135 of the signal processing unit 13 under the control of the control unit 6 will be described. As in the description of the embodiment, window functions to be candidates set for each of the amplification units 131, 132,..., 135 are determined. A plurality of candidate window functions are determined by non-linear optimization at each frequency under predetermined conditions in the microphone array device 100 such as the distance between microphones. Then, at a frequency at which the number of local minima points of the directional characteristic of the sound pickup is the same among the window functions obtained for each frequency, the window function obtained by nonlinear optimization is set to one amplification unit Classified as a candidate for Then, among window function candidates set in one amplification unit, a target directivity characteristic, for example, a window function having a main lobe width close to the target width is selected from the window functions as candidates, and this is selected. It is determined as a window function to be set in the amplification unit. Then, with respect to other window function candidates related to directional characteristics with different numbers of local minimum points, the window functions to be set to one amplification unit are determined in the same manner. Each of the microphones 9-1, 9-2, ..., 9-n in the present modification is a nondirectional microphone, but may be a microphone having directional characteristics. In this case, the window function may be determined in accordance with the directivity characteristics of each of the microphones 9-1, 9-2, ..., 9-n. [0047] Next, the frequencies f1, f2, f3, and f4 set in the signal division and synthesis unit 14 by the control of the control unit 6 will be described. As described in the embodiment, when sounds of various frequencies are picked up using the window function set in one amplification unit, the strength of the side lobes other than the main lobe increases in the high frequency band side, etc. Disturbance of the directional characteristics of Therefore, the upper limit frequency of each of the audio signals Sb-1, Sb-2, ..., Sb-4 output from each of the amplification units 131, 132, ..., 134 is set in a range where the disturbance of the directional characteristics does not occur. decide. Although there are various methods for determining the upper limit of the frequency, the upper limit of the frequency is determined within the range in which the intensity in a predetermined 09-05-2019 19 angular direction does not exceed a predetermined value. [0048] In this way, the microphone array device 100 can be made capable of performing sound collection with a predetermined directional characteristic, and the amount of calculation of signal processing can be reduced to achieve a wide range of frequencies, as in the effects of the embodiment. It becomes possible to perform directional control in a band by a process with low load. The modification 11 may apply a modification according to another modification to the microphone array apparatus 100. [0049] It is a block diagram showing composition of a speaker array device concerning an embodiment. It is a block diagram which shows the structure of the signal processing part which concerns on embodiment. It is a block diagram showing composition of an amplification part concerning an embodiment. It is a block diagram which shows the structure of the signal division part which concerns on embodiment. FIG. 7 is an explanatory view showing directivity characteristics in which the number of minimum points is 4 in directivity characteristics of an acoustic beam. It is explanatory drawing which shows the angle direction in the directional characteristic of an acoustic beam. It is explanatory drawing which shows the directional characteristic in which the number of the minimum points is 1 in the directional characteristic of an acoustic beam. FIG. 7 is an explanatory view showing directivity characteristics in which the number of minimum points is 2 in directivity characteristics of an acoustic beam. FIG. 7 is an explanatory view showing directivity characteristics in which the number of minimum points is 3 in directivity characteristics of an acoustic beam. It is explanatory drawing which shows the state in which the directivity of the acoustic beam was disturbed. It is explanatory drawing which shows the directional characteristic in which the number of the minimum points is 1 in the directional characteristic of an acoustic beam. It is explanatory drawing which shows the directional characteristic in which the number of the minimum points is 1 in the directional characteristic of an acoustic beam. It is a block diagram which shows the structure of the sound emission part which concerns on the modification 2. FIG. FIG. 18 is an explanatory view showing directivity characteristics in which the number of minimum points is 1 in the directivity characteristics of an acoustic beam according to Modification 3; FIG. 16 is a block diagram showing a configuration of an amplification unit according to a fifth modification. FIG. 16 is a block diagram showing a configuration of connection of a signal division unit, a delay circuit, and a signal processing unit according to a fifth modification. FIG. 18 is a block diagram showing a configuration of an 09-05-2019 20 amplification unit according to a modification 6; FIG. 18 is a block diagram showing the configuration of a signal processing unit according to Modification 7; FIG. 18 is a block diagram showing the configuration of a microphone array device according to a modification 11; FIG. 18 is a block diagram showing the configuration of a signal processing unit according to Modification 11. FIG. 16 is a block diagram showing the configuration of an amplification unit according to a modification 11; FIG. 18 is a block diagram showing the configuration of a signal division and synthesis unit according to Modification 11. Explanation of sign [0050] DESCRIPTION OF SYMBOLS 1 ... Speaker array apparatus, 2 ... sound emission part, 2-1 to 2-n ... Speaker, 21 ... Delay part, 21-1 to 21-n, 310, 320, 330, 340, 350-1 to 350-n ... Delay circuit 3, 13 ... Signal processing unit, 30, 1310, 140 ... Adder, 31-35, 131 to 135 ... Amplification unit, 31-1 to 31-n, 35-1 to 35-n, 131- 1-131-n ... Amplifier circuit, 4 ... Signal division unit, 4-1, 14-1 ... LPF, 4-2 to 4-4, 14-2 to 14-4 ... BPF, 4-5, 14-5 ... HPF, 5 ... Signal input unit, 6 ... Control unit, 7 ... Operation unit, 8 ... Storage unit, 9 ... Sound collection unit, 9-1 to 9-n ... Microphone, 14 ... Signal division synthesis unit, 15 ... Signal Output unit 09-05-2019 21
1/--страниц