JP2692817B2 - How to measure sound image resolution - Google Patents

Description

TECHNICAL FIELD The present invention relates to a method for measuring sound image resolution, that is, a method for measuring various characteristics such as sound image localization in stereo sound. [Prior Art] Sound reproduction is incomplete systematically, and even if a sound is recorded in a room, the same sound state cannot be reproduced. Further, it is said that the acoustic signal has a strong temporal change, and therefore a system for faithfully reproducing it cannot be configured. Furthermore, in the concert hall where you can directly listen to live sound, there are too many factors that cannot be controlled only by the magnitude of the sound pressure level distribution and the reverberation characteristics. Moreover, the correction effect that comes from vision cannot be ignored. Therefore, there is an element that cannot be understood unless a reproduction sound field for this purpose is created. On the other hand, of course, many physical characteristics of audio-related characteristics have been measured, and great effects have been achieved. [Problems to be Solved by the Invention] However, as described above, both the recording / reproducing system and the concert hall have system imperfections or unclear points. For this reason, all audio characteristics are incomplete by measuring and evaluating various physical characteristics, and the final evaluation system, that is, subjective evaluation by the human auditory / cognitive system is important together with the above-mentioned objective (physical characteristics) evaluation. There is. However, this subjective evaluation is greatly influenced by human preference, performance artistry, degree of auditory / cognitive training, individual differences, and the like. Therefore, in order to improve the audio evaluation, it is important to support the evaluation items that depend on the subjective evaluation with the objective evaluation. For example, the localization of a sound image, which is the main object of the present invention, that is, the sound image resolving power, conventionally depends only on subjective evaluation. In a live sound field, the visual / cognitive system works in conjunction with the auditory / cognitive system, so the sense of localization does not pose a problem, but in the audio visual reproduction, which is one form of reproduction, the screen size and speaker It raises basic problems such as the setting of. Furthermore, the surround system, which is one of the recent trends, utilizes human's sound image resolution, but it is still in the initial state from the viewpoint of reproduction close to the original sound. As described above, conventionally, the evaluation of the audio characteristic is incomplete, and in particular, the sound image localization, which is the origin of the stereo characteristic, that is, the sound image resolving power evaluation depends on the subjective evaluation of the human. Also, the basic data required for audiovisual systems and surround systems are lacking. For example, a live concert hall has almost no acoustic information by direction. An object of the present invention is to create a system for objectively measuring and evaluating sound image resolution as a physical property, which has hitherto relied on subjective evaluation. [Means for Solving Problems] Under such a purpose, in the measuring method of the present invention,
By providing a sound source and a sound collecting unit in a recording system, a sound generating unit in a reproducing system and a sound collecting unit assuming a listening range, and sequentially generating a plurality of frequency signals having different frequencies from the sound source in the recording system. , The sound pressure of the signal obtained by the sound collecting unit of the reproducing system and the phase thereof are measured for each frequency signal through the sound collecting unit of the recording system and the sound producing unit of the reproducing system, The sound image resolution through the system and the reproduction system is measured. [Operation] By the method as described above, it becomes possible to grasp various acoustic characteristics in the sound field, that is, frequency characteristics, delays, direction characteristics, sound pressure attenuation characteristics, reverberation characteristics, etc. at arbitrary points in the sound field. [Examples] Examples of the present invention will be described below. 1 to 8 are views for explaining an embodiment of the present invention. First, FIG. 2 is a diagram of a system based on this embodiment. In the recording space, the space between the sound source 1 and the microphone 2 is a medium for propagating sound waves, and the signal transmission medium is a delay circuit according to the distance, Acts as a damping circuit. Similarly, in the reproduction space, the space between the speaker and the listening point 4 also has the same function. FIG. 3 is a timing chart that schematically shows the states of the four stages of signals shown in FIG. When information is emitted from a sound source at time t ₀ , it propagates through the space and arrives at the microphone at the timing of t ₁ . On the other hand, when the time when information is electrically processed in the reproduction space and the information is inserted from the speaker into the reproduction space is different in real time, it is considered as t ₁ again,
The listening point is reached at the timing of t ₂ . In the present embodiment, a microphone is placed at this listening point to record this sound. More specifically, the flow of information at each stage at the time of recording, at the time of replay measurement, and at each stage is also determined by the multichannel. It is intended for digital recording. Further, FIG. 4 is a diagram for explaining the arrangement of the measuring instrument of the present embodiment, which uses the delay and attenuation effects brought about by the above-mentioned sound field space in the measurement of the sound image resolving power. First, in the recording space, a plurality of sound sources S1 to S5 (ideally omnidirectional single phase) are placed in the space, and a pair of microphones Ml, Mr ( Unidirectional) set toward the front. In the case of the sound source, in this example, the speaker S ₁ is _first placed at the apex of an equilateral triangle having a side length of d, and further the speaker S ₂ is located only d from the speaker S ₁ and d from the left microphone Ml. , Speaker S ₁ , S ₂
A speaker S _{3 at} a point d away from the speaker, a speaker S _{4 at} a point d away from the speakers S ₁ and S ₃ , and a speaker S ₅ at a point d away from the right microphone Mr and the speakers S ₁ and S ₄ , respectively. did. On the other hand, in the reproduction space, the speakers SPl and SPr are arranged at a distance d (center). On the other hand, it conformed to the IEC recommended conditions such as the microphone placement angle and listening points L1, L2, L3. However, other than the listening point L1 at the center, that is, the listening point L2 and the listening point L3
In order to see the stereo service area, in this example, it was set apart from the central listening point L1 by d / 2. Each listening point was focused on the left and right speakers. Two unidirectional microphones were set each. Next, the configuration of the entire measuring system of the present embodiment including the measuring devices arranged in this way will be described. 11, 12, 13, 1 in the figure
4,15, are control circuits, respectively, and the oscillators 21, 22, 23, 2 in the latter stage
Controls the oscillation timing and oscillation frequency of 4,25. Three
1,32,33,34,35 are amplifiers for amplifying the output signals of the oscillators 21,22,23,24,25, respectively, and the outputs thereof are the speakers S1, S2, S3, S4 arranged as described above. , S5 respectively. The control circuits 11, 12, 13, 14, and 15 are each equipped with a system controller 16
Data indicating the presence or absence of oscillation and the oscillation frequency is supplied from the control data generation circuit 17 on the basis of the control data. Further, 18 is an adder for adding the output signals of the amplifiers 31, 32, 33, 34, 35, 51 is an analog-digital (A / D) converter, 10
Is a multi-channel digital signal recording / reproducing apparatus. With the above configuration, first, the system controller 16 is an oscillator.
When 21 to 25 oscillate at regular intervals for a predetermined period, the control data generating circuits 11 to 15 are operated, the switches 26 and 27 are switched to the R side, and the device 10 is set to the recording state. Then, the switch 36 is connected to the A side, and the sound source signal obtained from the adder 18 and the signal obtained from the microphone Mr, that is, the collected sound signal are respectively converted into A / D converters 51 and 52 and switches 26 and 27. Through head 2
Multichannel recording at 8,29. Next, the same multi-channel recording is performed by connecting the switch 36 to the B side, and this time, the sound source signal and the signal obtained from the microphone Ml are similarly recorded. Recording on the above microphones Ml and Mr is performed for various frequencies by sequentially switching the frequencies of the sound source signals. Here, the deviation of the oscillation timing of the oscillators 21 to 25,
It is determined according to the assumed system, but in this example, it is set to 20 ms. The oscillation period of each oscillator 21 to 25 is set to a period longer than the wavelength of the oscillated signal. Also, A /
It is assumed that the D51 and A / D52 include not only the function of performing A / D conversion but also the function of performing digital signal processing for recording. Next, the operation of the portion corresponding to the reproduction system will be described.
Now, on the recording medium Ta, the sound source signals generated by the speakers S1 to S5 and the sound collection signals collected by the microphones Ml, Mr are recorded in parallel, but these heads 28, 29 are reproduced respectively. To be done. At this time, the switches 26 and 27 are connected to the P side by the control signals t1 and t2 from the system controller 16. Then, the reproduced sound collection signal reproduced from the head 29 is the speaker (SPr) corresponding to the microphone collecting the signal.
Alternatively, it is input to SPl) via the D / A converter 53. On the other hand, the system controller 16 connects the switches 41, 42, 43 to the R side by the control signals t5, t6, t7, respectively, and sets the multi-track digital recording / reproducing apparatus 20 in the recording mode. Now, assuming that the connection is made to the A side by the control signal t4 from the system controller 16, five kinds of signals generated from the speakers S1 to S5 and collected by the microphone Mr are sequentially supplied to the speaker SPr and reproduced. This reproduced signal is collected by one of the six microphones L1r, L1l, L2r, L2l, L3r, L3l arranged at the listening points L1, L2, L3. One of these six microphones (eg L1
r) is activated by the control signals t8 to t13 from the system controller, and the sound collection signal is digitally recorded on the recording medium Tb by the head 58 through the A / D converter 54 and the R side terminal of the switch 43. It On the other hand, the sound source signal reproduced by the heads 28 and 29 and the sound collection signal from the microphone Mr are supplied to the heads 56 and 57 through the delay circuits (DL) 54 and 55, and are output together with the output signal of the A / D converter 54 to the multi-channel. Will be recorded.
The delay times of the DLs 54 and 55 are the A / D converter 54 and the D / A converter 53.
It corresponds to the time required for various digital signal processing performed in. Next, this operation is performed for the five types of signals collected by the microphone Ml, and a series of these operations is performed for the sound source of each frequency. Recording of the sound collection signal for the microphone L1r is completed in this way, but the series of operations described above is performed for each microphone L1r.
Repeat for l, L2r, L2l, L3r, L3l. And finally, the control signal from the system controller 16
The switches 41, 42, and 43 are connected to the P side by t5, t6, and t7, respectively, so that the device 20 is in the reproduction state. With this, head 56,5
Reference numerals 7, 58 respectively reproduce the sound source signal, the sound collecting signal of the recording system, and the sound collecting signal at each listening point of the reproducing system, and supply it to the central processing unit (CPU) 50 as digital information. In the CPU50, input the signal reproduced as described above,
The sound source information, the sound collection information of the recording system, and the sound collection information of the reproduction system are analyzed for each frequency, sound collection microphone, and other factors, and stored as information that is easy to use for setting a concert hall or the like later. In the above configuration, the sound pickup signals of the microphones Ml and Mr are recorded in parallel by multitrack and reproduced by the speakers SPl and SPr in the same manner, and the six microphones L1l, L1r, L2l, L2r, It is an option in the configuration of the measuring device to allow simultaneous recording of sounds by L3l and L3r and multitrack recording of these in parallel with six heads, and should be allowed. In this way, one multichannel digital recording medium can obtain information on three stages: sound source, collected sound, and collected reproduced sound on the same time scale. What is important here is that the propagation velocity of the sound wave (about 340m / sec or 0.34m / ms) is overwhelmingly slower than the electrical signal propagation velocity when the test sound is generated at the sound source. It works as a delay circuit. Therefore, if the arrival of sound waves is measured at the 1/100 millisecond level, the distance to the sound source can be measured with an accuracy of 3 mm. FIG. 5 shows the omnidirectional microphones Ml and M1 that emit burst signals from the test sound sources S1 to S5.
Speakers that correspond to the signals collected and recorded by Mr.
Playback from SPl, SPr, directional microphones L1l, L1r, L2l, L2 facing each speaker at three measurement points L1-L3
The data collected by r, L3l, and L3r are shown schematically.
However, the microphones L1r, L2r, and L3r show data on the reproduced sound of the speaker SPr, and the microphones L1l, L2l, and L3l show data on the reproduced sound of the speaker Spl. It can be seen that if there is no phase error in the reproduction system, sound waves arrive at each listening point (two measurement points) with a delay according to the position of the sound source and the propagation path, as shown in the figure. The energy of an actual sound wave is attenuated as it propagates.
As is well known, in the free sound field, it is inversely proportional to the square of the distance, and indoors, etc., it is approximately inversely proportional although it depends on the conditions. For the purpose of explaining both the recorded sound field and the reproduced sound field, let us consider an anechoic room, that is, a free sound field. Of course, this method is effective even in a normal sound field. In this case, when the distance is doubled, the sound pressure becomes 1 /
4, ie 12 dB or less. Information on the sound source position can be obtained from these characteristics, that is, the transmission direction, phase (timing), and sound pressure of the sound wave.
Conversely, depending on the characteristics of the recording and reproducing system, the sound waves reaching L1 to L3 may have distortion, phase shift, and variation in sound pressure, which naturally causes variations in measured values. When the sound source is obtained by back-calculating from the data including such variations, there is naturally no single point and the spread occurs. By changing the frequency of the test sound, the original sound image corresponding to each frequency can be obtained. As a result, it is possible to illustrate how the phase shift, the sound pressure unevenness, and the like due to the crossover occurring in the multi-way speaker system impair the stereo feeling. Further, conventionally, there is an intensity stereo that gives a stereo feeling only by a so-called level difference in that the balance of the sound pressure has the greatest effect on the stereo feeling and the phase shift has a slight effect. In addition, the diffuse reflection stereo of the famous US speaker maker BOSE is the company's representative work BOSE90.
As can be seen in 1, the direct and indirect sounds are mixed together and the phase information cannot be persistently obtained, and a stereo feeling is obtained only by the level difference. In other words, Intensinai Stereo cuts off the phase information when creating a recording media, and BOSE type when playing. However, this kind of approach is already known to be inconsistent with modern psychoacoustics. According to the above-described embodiment, since the sound image resolution due to the sound pressure level difference and the sound image resolution due to the phase difference can be evaluated independently,
It is also possible to elucidate the relationship between the sensory and subjective scales of so-called stereo feeling and the physical and objective scales of sound image resolution. As this kind of clarification progresses, the standardization of the sound field for recording and reproduction and the positional distance relationship such as the setting of microphones and speakers will be promoted, and reference information can be obtained on how to apply a delay even with a multi-microphone. The origin of the present invention was not to rely on human subjective evaluation, but to perform localization of sound sources by physical measurement of information provided by the reproduction system, that is, by audience evaluation. This is the area of the transfer function corresponding to the resolving power of the camera / lens which is the same information reproducing system as described above. However, as evidenced by the evaluation of physical quantities, the various characteristics that the optical system captures with human sense,
Also in the case of the acoustic system, it is important to compare the sensory quantity with the physical quantity. The optical system has a visibility curve, and the acoustic system has a hearing curve. Each of them is an important characteristic that connects physical quantity and sensory quantity. Just as the color optical device is designed according to the human color discrimination ability, the method of designing the audio device in consideration of the auditory characteristics is effective. For example, human sound image localization force depends on frequency,
The localization power decreases at low and high temperatures. This is not unrelated to the significant decrease in low temperature and high temperature cosensitivity, as is apparent from the audiometric curve in FIG. Conversely, when an error of the same physical quantity is unavoidable, the physical characteristics corresponding to the sense quantity are improved by distributing the same in a region where human sensitivity is low.
That is, the sound image resolution in the middle range should be given the highest priority. The sound image resolution measuring system described above can also be used for characterizing the sound field, especially in concert halls and surround systems, by adopting a partially modified arrangement. FIG. 6 shows a conceptual diagram of a concert hall sound field characteristic measuring system. 5 is a microphone system for measurement. As shown in the lower right part of the figure, 6 microphone sets are provided in a wide shape toward the front, 2 at the left, 2 at the left, 1 at the top and 1 at the rear. The two to the front have a directivity of 45 °, and the other has a unidirectional directivity. In addition, Sl and Sr are speakers, which are unidirectional and have no single phase. Next, referring to this figure, the purpose of this embodiment will be described. First, a test sound is emitted from the two speakers Sl and Sr. These test sounds are collected by the microphone system 5 placed at the listening point, and the original sounds are recorded together with the original sound in the digital multi-channel recorder by the system as shown in FIG. FIG. 7 is a diagram schematically showing the relationship between the sound source signal and the sound collection signal according to the arrangement shown in FIG. 6. First, as shown in the figure, the sound directly arrives at Fl and Fr facing the front left and right. Next, the primary reflected sound reflected on the back surface is attenuated a little and arrives. On the other hand, the sounds reflected by the side surfaces arrive at the left and right microphones L and R, respectively, with a delay of the distance between the left and right speakers. In the rear microphone B, the weak input that came back by hitting the opposing surface of the stage enters with a delay.
Also, the sound reflected from the ceiling enters the upper microphone T. Of course, sound waves have a large diffraction phenomenon, unlike light,
Not only the linear propagation shown in FIG. 6, but in general, there are many commonalities with light such as reflection, absorption, and interference of waves.
Further, in FIG. 7, only a part of the first wave and the second wave are shown for the purpose of explaining the principle, but in addition to this, there are innumerable inputs that form so-called reverberation sounds such as multi-order reflection and diffraction input, The integrated value of sound pressure shows exponential decay with respect to the time axis. The features of the system of the above embodiment are as follows. i) Conventionally, the reverberation characteristic was measured by input from all directions, but this method can be measured by the direction of arrival of sound. ii) By making the input to the speaker a past wave,
The main input for each direction is obtained as a pulse for each route, and the sound wave path can be clarified by comparison with the internal structure of the concert hall. Here, too, the propagation speed of the sound wave is an important factor as a delay circuit. iii) By changing the frequency of the input to the speaker,
It is possible to understand the sound field characteristics that result from the difference in reflection, absorption, and diffraction of sound depending on the frequency. iv) The sound pressure level distribution and its factors can be understood by changing the setting location of the microphone system. This system is used to evaluate concert halls and other performance venues.
It is extremely effective for improvement. In addition, the charm and secrets of a famous concert hall can be clarified. This data is indispensable for enjoying the atmosphere of a concert hall in the world, which is a wish of classical music lovers. The reverberation created by the mixture of reflected sounds controlled from all over the space is always a small live space for homes and the usual 2
It cannot be reproduced with the reverberant sound from the speaker stereo. After all, in order to get the feeling of a large hall at home, in addition to the previous two, left and right,
The surround system, in which speakers are placed above and behind, is not accepted. However, the current surround system uses a synthetic output and a delay circuit different from each other for each manufacturer except for the previous two sound sources, and the system is still under development. The system of the above-mentioned embodiment provides various data necessary for reproducing the reproduced sound field as close to the original sound field as possible, that is, detailed acoustic information for each direction. The acoustic information mentioned here refers to direct waves and indirect waves (primary reflection, secondary reflection, etc.), the structure of each concert hall, the direction specified from the listening point, the attenuation characteristics by frequency of time-series input, etc. Pointing to. Because these characteristics are different,
Special seats at Carnegie Hall or Concerto Bo,
The presence of the box seats is different. Therefore, at home, in order to obtain the presence of a well-known concert hall, that is, the reproduced sound field close to the original sound field, the direction, the right and left, corresponding to the attenuation characteristics for each direction obtained by the above measurement and for each frequency of time series input, Input conditions to the upward and backward speakers are combined using a delay circuit and frequency-dependent attenuation filter,
Combined with a normal stereo playback speaker, a surround system is created. Ideally, each direction is recorded with multi-channel,
It is desirable to reproduce from a corresponding speaker, but even with the above-described synthesis method, a realistic sensation close to that can be obtained. In order to operate this system effectively, it is better that both the reverberation characteristic at the time of recording and the reverberation characteristic of the reproduced sound field are slightly distorted. This is because the surround system to which the above method is applied can be considered as an omnidirectional reverberation applying system. In addition, the delay and direction characteristics corresponding to a large space are important for reproducing the sound field of large theaters and movie theaters, which is another purpose of the surround system. It is practically effective to store these conditions in an internal or external memory so that they can be used for any purpose, and that each concert hall can be easily reproduced. Also, it is good to set the home position as desired. In this system as well, it is desirable to design the system taking into consideration not only the objective and physical characteristics but also the human sensitivity curve, as with the sound image resolution described above. For example, when the azimuth dependence of the reverberation in a concert hall becomes clear, a surround system that reproduces it relies on the azimuth-sensitive midrange from the corresponding speakers, and the low and high ranges are also used as the main speakers. I can't accept it. Up to this point, it has already been partially put to practical use, but this method also takes into account the reverberation characteristics of the reproduced sound field. That is, it is desirable that the reproduced sound field is basically dead, that is, has less reverberation for surround reproduction. In particular, the reverberation in the middle range should be short. Otherwise, even if the sound is reproduced aiming at the original sound field from each direction, the reverberation will act as a mask, impairing the sense of direction and the sense of presence. Moreover, it is also undesirable that the recording itself contains extremely whole tones. With the system as described above, it has been more than 30 years since the transition from monaural to stereo that the sense of stereo, that is, the sound image resolution, can be grasped as a physical characteristic. This sound image resolution corresponds to the resolution of a camera / lens which is the same information reproducing system. Therefore, the evaluation method can be either the resolution method or the MTF method, depending on the application. Further, the sound image resolution also varies in the sound field, as the resolution of the lens differs between the main axis and the periphery. Therefore, the angle of view corresponds to the stereo reproduction range, and the distribution of resolving power in that range is naturally symmetrical for measurement and evaluation. Furthermore, the present invention makes it possible to understand various acoustic characteristics in a live sound field, that is, frequency characteristics, delays, direction characteristics, sound pressure level, reverberation characteristics, etc. at various points in the sound field. Furthermore, the present invention can achieve the following objects by grasping the acoustic characteristics of both the live sound field and the reproduced sound field. i) Standardization of recording conditions, especially microphone setting and reproduction conditions, especially speaker arrangement, etc., enables complete reproduction viewed from the sound image. ii) In the audio playback system and surround playback system, it is possible to play back a sound field closer to the original sound field. iii) Various audio equipment from the entrance to the exit. Also, the sound image resolution of the entire system can be evaluated independently or in combination. By the way, the main reasons why such a system has become possible are described below. i) Advances in digital audio technology have enabled multichannel digital recording. ii) By improving the quality of recording and playback systems, it became possible to apply the transfer function and theory used in optical systems to sound wave propagation in a sound field. iii) Improvement of computer technology. On the other hand, improvements in recording and playback systems have reached the area where it is difficult to find differences in sound image localization based on subjective evaluation.
It is possible to view that they are seeking to improve physical and objective evaluation. [Effects of the Invention] As described above, the present invention provides measurement of audio characteristics,
Evaluation In particular, the following effects are obtained by measuring and evaluating the sound image resolution of the reproduction system. i) The sound image resolving power of the reproduction system, which relies on the subjective evaluation, can be subjectively measured and evaluated as a physical characteristic. ii) Acoustic resolution can identify and evaluate any place in the sound field. iii) It is possible to completely reproduce the sound image position by standardizing the recording conditions, especially the setting of the sound collecting section. iv) The sound image resolving power of various audio equipment from the entrance to the exit and the entire system can be evaluated independently or in combination. Further, by applying the present invention to the sound field measurement and control of a concert hall, a surround system, etc., v) time series intensity and frequency characteristics can be obtained for each direction of arrival of sound waves, and conventionally, they have been collectively known as reverberation characteristics. Things can be analyzed by direction. vi) Obtain useful data for designing and improving the internal structure of the sound field. vii) Time-sequential reverberation characteristics by direction of a famous concert hall are measured and used to set the conditions when reproducing a sound field with a surround system, and the atmosphere of a special seat can be reproduced at home.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram showing a schematic configuration of a measurement system as an embodiment of the present invention, FIG. 2 is a diagram showing a system premised on the embodiment of the present invention, and FIG. FIG. 4 is a diagram schematically showing four stages of signals shown in FIG. 2, FIG. 4 is a diagram showing the arrangement of measuring instruments in the present embodiment, and FIG. 5 is a diagram showing signals of each part according to the arrangement of FIG. , FIG. 6 is a diagram showing another example of the arrangement of measuring instruments, FIG. 7 is a diagram schematically showing the signals of the respective parts according to the arrangement of FIG. 6, and FIG. 8 is a human hearing sensitivity characteristic. FIG. In the figure, S1, S2, S3, S4, S5, Sl, and Sr are speakers as sound sources, Ml,
Mr, 5 is a microphone, SP as a sound collection part in the recording system
l and SPr are speakers as a sound generator of the reproduction system, L1, L2, L3 are listening points, L1r, L1l, L2r, L2l, L3r and L3l are microphones as a sound collector of the reproduction system, and 10 and 20 are respectively A digital recording / reproducing apparatus, 16 is a system controller, and 50 is a CPU.

Claims (1)

(57) [Claims] By providing a sound source and a sound collecting section in the recording system, a sound generating section in the reproducing system and a sound collecting section assuming a listening range, and sequentially generating a plurality of frequency signals having different frequencies from the sound source in the recording system, the recording system The sound recording unit and the reproducing system by measuring the sound pressure and the phase of the signal obtained by the sound collecting unit of the reproducing system for each frequency signal via the sound collecting unit of the reproducing system and the sound generating unit of the reproducing system. A method for measuring sound image resolution, which comprises measuring the sound image resolution through the image. 2. The sound collecting unit of the recording system includes a plurality of sound collectors having different directivities, the sound generating unit includes a plurality of speakers, and the sound pressure and the phase are sequentially measured for each sound collector. A method for measuring a sound image resolving power according to claim (1).