JP6003355B2 - Reverberation time analyzer - Google Patents

Description

  The present invention relates to a technique for measuring the reverberation time of an acoustic space.

  A technique for measuring reverberation time in an acoustic space has been proposed (for example, Patent Document 1). For example, a reverberation curve (attenuation curve) is calculated by an impulse integration method using a specific extraction interval as an integration range among impulse responses measured in an acoustic space, and a time length during which the signal intensity attenuates by a predetermined amount in the reverberation curve is calculated. There is a technique for calculating reverberation time.

JP 2005-12784 A

  However, in the impulse integration method, the shape of the reverberation curve varies depending on the time length of the extraction interval that is the integration range of the impulse response in the acoustic space, so if the time length of the extraction interval is not properly selected, the reverberation time is measured. There is a problem that accuracy decreases. In view of the above circumstances, an object of the present invention is to specify the reverberation time with high accuracy from the impulse response of the acoustic space.

  In order to solve the above problems, a reverberation time analysis apparatus according to the present invention includes a first analysis unit that specifies a first reverberation curve corresponding to a first extraction section of an impulse response in an acoustic space, A reverberation time provisional means for calculating a provisional reverberation time according to a time when the intensity is attenuated by a predetermined amount, and a second reverberation curve corresponding to a second extraction interval over a time length corresponding to the provisional reverberation time among impulse responses are specified. Second analysis means and reverberation time calculation means for calculating reverberation time from the second reverberation curve are provided. In the above configuration, the provisional reverberation time is calculated from the first reverberation curve corresponding to the first extraction interval of the impulse response, and from the reverberation curve corresponding to the second extraction interval of the time length corresponding to the provisional reverberation time of the impulse response. Reverberation time is calculated. Therefore, there is an advantage that the reverberation time can be specified with high accuracy by reducing the influence of the length of the extraction interval set in the impulse response.

  In a preferred aspect of the present invention, the second analysis means sets a second extraction interval over a time length obtained by multiplying the provisional reverberation time by a predetermined value (for example, coefficient κ). According to the above aspect, there exists an advantage that the time length of a 2nd extraction area can be set easily.

  In a preferred aspect of the present invention, the reverberation time provisional means sets the provisional reverberation time according to the length of time that the signal intensity attenuates by a predetermined amount immediately after the attenuation start point in the first reverberation curve. Since the section immediately after the attenuation start point in the first reverberation curve approximates an ideal reverberation curve in which the time length of the integration range is set to the optimum value, the signal intensity immediately after the attenuation start point in the first reverberation curve. According to the configuration in which the provisional reverberation time is set according to the length of time during which the sound is attenuated by a predetermined amount, the effect that the reverberation time can be specified with high accuracy is particularly remarkable.

  The reverberation time analysis apparatus according to each of the above aspects is realized by hardware (electronic circuit) such as a DSP (Digital Signal Processor) dedicated to reverberation time analysis, and a general-purpose such as a CPU (Central Processing Unit). This is also realized by cooperation between the arithmetic processing unit and the program. The program according to the present invention includes a first analysis process for identifying a first reverberation curve corresponding to a first extraction interval of an impulse response of an acoustic space, and a time during which the signal intensity attenuates by a predetermined amount in the first reverberation curve. From the reverberation time provisional process for calculating the provisional reverberation time, the second analysis process for specifying the second reverberation curve corresponding to the second extraction section over the time length corresponding to the provisional reverberation time in the impulse response, and the second reverberation curve A computer executes a reverberation time calculation process for calculating a reverberation time. According to the above program, the same operation and effect as the reverberation time analyzing apparatus according to the present invention are realized. The program according to the present invention is provided in a form stored in a computer-readable recording medium and installed in the computer, or provided in a form distributed via a communication network and installed in the computer.

1 is a block diagram of an acoustic characteristic evaluation device according to a first embodiment of the present invention. It is a flowchart which shows the procedure of liveness evaluation. It is a block diagram of a 1st characteristic analysis part. It is explanatory drawing of the reverberation curve specified from an impulse response. It is a graph which shows the relationship between an average sound absorption rate and the basic value for calculation of a 1st index value. It is a schematic diagram of an evaluation screen. It is a schematic diagram of a reverberation characteristic screen. It is a schematic diagram of a sound absorption characteristic screen. It is a flowchart which shows the procedure of flutter echo evaluation. It is a block diagram of the 2nd characteristic analysis part. It is an example of a display of the result of flutter echo evaluation. It is an example of a display of the result of flutter echo evaluation. It is a flowchart which shows the procedure of booming evaluation. It is explanatory drawing of the relationship between the sound collection point at the time of booming evaluation, and a sound emission point. It is an example of a display of the result of booming evaluation. It is an example of a display of the result of booming evaluation. It is a block diagram of the reverberation time analysis part in 2nd Embodiment. It is explanatory drawing of operation | movement of the reverberation time analysis part in 2nd Embodiment. It is a graph which shows the relationship between the average sound absorption rate in a modification, and the basic value for calculation of a 1st index value. It is a block diagram of the 2nd characteristic analysis part concerning a modification.

<First Embodiment>
FIG. 1 is a block diagram of an acoustic characteristic evaluation apparatus 100 according to the first embodiment of the present invention. The acoustic characteristic evaluation apparatus 100 is an apparatus that evaluates the acoustic characteristics of an arbitrary acoustic space. As shown in FIG. 1, the arithmetic processing apparatus 10, the storage device 12, the display device 14, the input device 15, and the sound collection device 16. It is implement | achieved by the computer system which comprises. Specifically, the acoustic characteristic evaluation apparatus 100 can be realized by a portable information processing apparatus such as a mobile phone or a smartphone.

  The storage device 12 stores a program PGM executed by the arithmetic processing device 10 and various data used by the arithmetic processing device 10. A known recording medium such as a semiconductor recording medium or a magnetic recording medium or a combination of a plurality of types of recording media can be arbitrarily employed as the storage device 12.

  The display device 14 (for example, a liquid crystal display panel) displays various images under the control of the arithmetic processing device 10. For example, the evaluation result of the acoustic characteristics in the acoustic space is displayed on the display device 14. The input device 15 is a device for a user to input an instruction to the acoustic characteristic evaluation device 100, and includes a plurality of operators operated by the user, for example. Note that a touch panel configured integrally with the display device 14 may be employed as the input device 15.

  The sound collection device 16 is a microphone that generates an observation signal Y by collecting surrounding sounds in the acoustic space. The sound collection device 16 is typically built in the acoustic characteristic evaluation device 100, but a configuration in which the sound collection device 16 configured separately from the acoustic property evaluation device 100 is connected to the acoustic characteristic evaluation device 100 is also possible. Can be employed. The A / D converter that converts the observation signal Y generated by the sound collection device 16 from analog to digital is not shown for convenience.

  A sound emitting device 17 (speaker) is connected to the acoustic characteristic evaluation device 100. An acoustic signal (hereinafter referred to as “measurement signal”) X for measuring acoustic characteristics in the acoustic space is supplied from the acoustic characteristic evaluation device 100 to the sound emitting device 17. The sound emitting device 17 radiates sound corresponding to the measurement signal X supplied from the acoustic characteristic evaluation device 100 (hereinafter referred to as “measurement sound”) into the acoustic space. The D / A converter that converts the measurement signal X from digital to analog is not shown for convenience.

  The arithmetic processing device 10 executes a program PGM stored in the storage device 12 to thereby evaluate a plurality of functions (acoustic reproduction unit 20, signal acquisition unit 22, response acquisition unit 24, The analysis processing unit 26 and the display control unit 28) are realized. A configuration in which the functions of the arithmetic processing device 10 are distributed to a plurality of devices, or a configuration in which a dedicated electronic circuit (DSP) realizes a part of the functions of the arithmetic processing device 10 may be employed.

  The sound reproducing unit 20 generates the measurement signal X and supplies it to the sound emitting device 17 so that the measurement sound is emitted from the sound emitting device 17 into the acoustic space for a predetermined time. The measurement signal X is an acoustic signal suitable for measuring the impulse response of the acoustic space. Specifically, a noise signal such as a time stretched signal (TSP: Time Stretched Pulse) whose frequency continuously changes in time or pink noise whose signal intensity is distributed over a wide band is preferably used as the measurement signal X.

  The signal acquisition unit 22 acquires the observation signal Y from the sound collection device 16 in synchronization with the supply of the measurement signal X (radiation of measurement sound) by the sound reproduction unit 20. Specifically, the signal acquisition unit 22 acquires the observation signal Y over a predetermined time (a time length sufficient for analysis of the acoustic characteristics of the acoustic space) from the time when the supply of the measurement signal X to the sound emitting device 17 stops. To do. The measurement sound radiated from the sound emitting device 17 arrives at the sound collecting device 16 through reflection and scattering on the wall surface of the acoustic space. Accordingly, the observation signal Y acquired by the signal acquisition unit 22 reflects the acoustic characteristics (reverberation sound) of the acoustic space.

  The response acquisition unit 24 specifies the impulse response R of the acoustic space from the observation signal Y acquired by the signal acquisition unit 22. For the identification of the impulse response R by the response acquisition unit 24, a known calculation process according to the type of the measurement sound measurement signal X is used.

  The analysis processing unit 26 analyzes the acoustic characteristics of the acoustic space using the impulse response R acquired by the response acquisition unit 24. The analysis processing unit 26 according to the first embodiment includes a first characteristic analysis unit 261, a second characteristic analysis unit 262, and a third characteristic analysis unit 263. The first characteristic analysis unit 261 evaluates the liveness (reverberation) of the acoustic space. The second characteristic analysis unit 262 evaluates flutter echo in the acoustic space. Flutter echo is a phenomenon in which reverberant sound is periodically perceived by a listener in the acoustic space due to repeated reflection of sound between the inner wall surfaces of the acoustic space. The third characteristic analysis unit 263 evaluates booming in the acoustic space. Booming is a phenomenon in which a specific band component (typically a low frequency range) of sound is emphasized by a standing wave (resonance) in an acoustic space, and is perceived as unnatural sound. The display control unit 28 causes the display device 14 to display the result of analysis by the analysis processing unit 26.

  The user can arbitrarily select any of liveness, flutter echo, and booming as an evaluation target by appropriately operating the input device 15. The acoustic characteristic evaluation apparatus 100 executes processing according to the evaluation target selected by the user. Specific operation of the acoustic characteristic evaluation apparatus 100 will be described in detail below for each evaluation target.

<Liveness evaluation>
FIG. 2 is a flowchart of a procedure for evaluating liveness of an acoustic space. The operation of FIG. 2 is started in response to an instruction from the user to the input device 15 (instruction to start liveness evaluation). The sound emitting device 17 is disposed at a predetermined position (for example, a corner) in the acoustic space.

  When the process of FIG. 2 starts, a volume adjustment process for adjusting the volume of the measurement sound used for measuring the acoustic characteristics is executed (SA1). Specifically, the sound reproduction unit 20 radiates the measurement sound from the sound emitting device 17 by supplying the measurement signal X, and the observation signal Y acquired by the signal acquisition unit 22 from the sound collection device 16 during the emission of the measurement sound. The signal intensity of the measurement signal X (the volume of the measurement sound) is adjusted so that the signal intensity becomes a numerical value within a predetermined range. It is also possible for the user to manually adjust the volume of the measurement sound by operating the input device 15.

  When the volume adjustment process is completed, impulse responses R [n] (R [1] to R [N]) are sequentially measured at each of N receiving points (N is a natural number of 2 or more) in the acoustic space. (SA2). Each point having a different positional relationship with respect to the sound emitting device 17 in the acoustic space is preferably selected as a sound receiving point.

  Specifically, the user moves the acoustic characteristic evaluation device 100 (sound collecting device 16) to each sound receiving point in the acoustic space, and instructs the input device 15 to measure the measurement sound at each sound receiving point. Give. When the measurement sound is instructed at the nth (n = 1 to N) sound receiving points, the sound reproducing unit 20 supplies the signal X for measuring the signal intensity set in the volume adjustment processing. The measurement sound is radiated from the sound emitting device 17 to the acoustic space for a predetermined time, and the signal acquisition unit 22 acquires the observation signal Y [n] for a predetermined time from the time when the supply of the measurement signal X stops from the sound collection device 16. To do. The response acquisition unit 24 calculates the impulse response R [n] of the acoustic space from the observation signal Y [n] acquired by the signal acquisition unit 22 and stores it in the storage device 12. By executing the above process sequentially at each of the N sound receiving points, N impulse responses R [1] to R [N] corresponding to different sound receiving points are stored in the storage device 12. Is done.

  When the measurement of the impulse response R [n] is completed for N sound receiving points in the acoustic space, the user operates the input device 15 to set the acoustic space size Z to the acoustic characteristic evaluation device 100 (analysis processing unit). 26) (SA3). For example, assuming that the acoustic space has a rectangular parallelepiped shape, the analysis processing unit 26 receives the height, width, and depth of the acoustic space as a size Z from the user. In addition, the structure which receives designation | designated of the size Z before measurement of each impulse response R [n], and the structure which acquires the size Z previously memorize | stored in the memory | storage device 12 can also be employ | adopted.

  When the measurement of the impulse response R [n] for each sound receiving point and the reception of the size Z of the acoustic space are completed by the above procedure, the first characteristic analysis unit 261 is an evaluation index that is a measure of the suitability of the liveness of the acoustic space. Calculate SL (SA4). The display control unit 28 causes the display device 14 to display the evaluation result (evaluation index SL) by the first characteristic analysis unit 261 (SA5).

  FIG. 3 is a block diagram of the first characteristic analysis unit 261. As shown in FIG. 3, the first characteristic analysis unit 261 includes a sound absorption characteristic analysis unit 32 and an evaluation processing unit 34. The sound absorption characteristic analysis unit 32 analyzes each impulse response R [n] for each sound receiving point, so that each of M bands (M is a natural number of 2 or more) on the frequency axis (hereinafter referred to as “evaluation band”). The average sound absorption coefficient α [m] (α [1] to α [M]) is calculated. The average sound absorption coefficient α [m] (m = 1 to M) corresponds to the rate at which the acoustic component of the mth evaluation band is absorbed (absorbed or transmitted) by the wall surface of the acoustic section. Each evaluation band is set to a predetermined bandwidth. For example, six (M = 6) evaluation bands are set by dividing a frequency band from 125 Hz to 4 kHz for each 1/1 octave band.

  As shown in FIG. 3, the sound absorption characteristic analysis unit 32 includes a band division unit 322, a reverberation time analysis unit 324, and a sound absorption rate calculation unit 326. The band dividing unit 322 performs a filter (for example, a bandpass filter) process on the impulse response R [n] of each sound receiving point, so that M band impulse responses RB [n] corresponding to different evaluation bands are obtained. , 1] to RB [n, M] are generated for each of the N sound receiving points. The band impulse response RB [n, m] is an acoustic component in the mth evaluation band of the impulse response R [n] measured at the nth sound receiving point in the acoustic space.

  The reverberation time analysis unit 324 analyzes each band impulse response RB [n, m] after being processed by the band dividing unit 322, so that M reverberation times T [1] to T [corresponding to different evaluation bands are obtained. M] is calculated. The reverberation time T [m] corresponds to the time length until the intensity of the acoustic component of the mth evaluation band decreases by 60 dB from the start of attenuation in the acoustic space. The reverberation time analysis unit 324 calculates the reverberation time T [m] according to the band impulse responses RB [1, m] to RB [N, m] of the N sound receiving points corresponding to the mth evaluation band. To do.

  The impulse integration method (Schroeder method) is preferably used for calculating the reverberation time T [m]. Specifically, the reverberation time analysis unit 324 performs the reverberation curve (attenuation curve) C [of FIG. 4 by an impulse integration method in which a predetermined interval on the time axis of each band impulse response RB [n, m] is an integration range. n, m] is calculated, and the reverberation curve CA [m] corresponding to the reverberation curves C [1, m] to C [N, m] of the N receiving points corresponding to the mth evaluation band is obtained. The reverberation time T [m] of the mth evaluation band is calculated. The reverberation curve CA [m] is, for example, an average (that is, an ensemble average) of N reverberation curves C [1, m] to C [N, m] corresponding to the mth evaluation band. The reverberation time analysis unit 324 calculates the reverberation time T [m] (T [1] to T [M]) for each evaluation band by executing the above process for each of the M evaluation bands. In the above example, the reverberation time T [m] is calculated from the reverberation curve CA [m] obtained by averaging N reverberation curves C [1, m] to C [N, m]. Reverberation time T [m] is calculated by averaging N reverberation times specified from each reverberation curve C [n, m] (C [1, m] to C [N, m]) corresponding to the band. It is also possible to do.

The sound absorption coefficient calculation unit 326 in FIG. 3 corresponds to different evaluation bands from the reverberation times T [1] to T [M] calculated by the reverberation time analysis unit 324 and the size Z of the acoustic space designated by the user. M average sound absorption coefficients α [1] to α [M] are calculated. Specifically, the sound absorption coefficient calculation unit 326 calculates the average sound absorption coefficient α [m] of the mth evaluation band by calculation of the following formula (1) (Eyling's reverberation expression).

  The symbol S in Equation (1) means the inner area of the acoustic space (the total area of the inner wall surface), and the symbol V means the volume of the acoustic space. The inner area S and the volume V are calculated from the size Z of the acoustic space designated by the user in step SA3 in FIG. The symbol K is a predetermined chamber constant (for example, K = 0.162). The above is the configuration and operation of the sound absorption characteristic analysis unit 32 in FIG.

  The evaluation processing unit 34 in FIG. 3 analyzes the average sound absorption rate α [m] (α [1] to α [M]) for each evaluation band calculated by the sound absorption characteristic analysis unit 32 (sound absorption rate calculation unit 326). To calculate the liveness evaluation index SL of the acoustic space. The evaluation index SL designates one of five levels (Excellent / Good / Fair / Poor / Bad) that categorize the liveness evaluation results. Specifically, the evaluation index SL is set to any value from a numerical value “1” corresponding to the lowest evaluation (Bad) to a numerical value “5” corresponding to the highest evaluation (Excellent). As shown in FIG. 3, the evaluation processing unit 34 includes a first evaluation unit 342, a second evaluation unit 344, and an index calculation unit 346.

  The first evaluation unit 342 calculates a first index value SL1 according to the M average sound absorption rates α [1] to α [M] calculated by the sound absorption characteristic analysis unit 32. The first index value SL1 is an index for evaluating the liveness of the acoustic space from the viewpoint of the magnitude of the absolute amount of sound absorption in the acoustic space (deviation from the optimum value). The first index value SL1 of the first embodiment is any numerical value from 1 (minimum evaluation) to 5 (maximum evaluation) corresponding to each stage (Excellent / Good / Fair / Poor / Bad) that classifies the evaluation results. Set to

  Specifically, the first evaluation unit 342 calculates a basic value σ [m] of each evaluation band by calculation using the average sound absorption coefficient α [m], and M basic values corresponding to different evaluation bands. A first index value SL1 corresponding to σ [1] to σ [M] is calculated. For example, it corresponds to a numerical value range including an average value of M basic values σ [1] to σ [M] among five numerical value ranges corresponding to the numerical values (1 to 5) of the first index value SL1. A numerical value (1-5) is calculated as the first index value SL1.

The basic value σ [m] of each evaluation band is calculated by, for example, the following equation (2) using the average sound absorption coefficient α [m].

The symbol abs [] in Equation (2) means an absolute value. The symbol α0 in the formula (2) is a predetermined value corresponding to the optimum value of the average sound absorption coefficient α [m], and is set to a numerical value of about 0.25, for example. Log ₁₀ (α [m] / α0) in parentheses in the formula (2) means a logarithmic distance (difference between logarithmic values) between the average sound absorption coefficient α [m] and a predetermined value α0. FIG. 5 is a graph showing the relationship between the average sound absorption coefficient α [m] and the basic value σ [m]. As understood from the equation (2) and FIG. 5, the basic value σ [m] becomes a larger numerical value as the difference (distance) between the average sound absorption coefficient α [m] and the predetermined value α0 is larger. As the basic value σ [m] of each evaluation band is smaller (the average sound absorption coefficient α [m] is closer to the predetermined value α0), the first index value SL1 is larger (numerical value indicating higher evaluation). The 1 evaluation unit 342 sets the first index value SL1.

  In the configuration for calculating the logarithmic distance between the average sound absorption coefficient α [m] and the predetermined value α0 as expressed by the mathematical formula (2), the average sound absorption coefficient α [m] is lower than the predetermined value α0, as can be seen from FIG. When the average sound absorption coefficient α [m] is changed by the unit amount within the range QL, the change amount of the basic value σ [m] is within the range QH where the average sound absorption coefficient α [m] exceeds the predetermined value α0. It exceeds the amount of change of the basic value σ [m] when α [m] changes by the unit amount. That is, in the range QL where the average sound absorption coefficient α [m] is low, the basic value σ [m] varies in conjunction with the average sound absorption coefficient α [m] compared to the range QH where the average sound absorption coefficient α [m] is high. Easy to do.

  In an acoustic space with a large amount of reverberation (a space with a low average sound absorption rate), the perceptual sensitivity to changes in the amount of reverberation is relatively high, and an auditory sense of incongruity tends to occur when the amount of reverberation increases or decreases. In space (a dead space with a high average sound absorption rate), the perceptual sensitivity to changes in the amount of reverberation is relatively low. Exists. In the above-described configuration for calculating the logarithmic distance between the average sound absorption coefficient α [m] and the predetermined value α0, the average sound absorption coefficient α [m] is compared with the range QH where the average sound absorption coefficient α [m] is high within the low range QL. Since the variation of the basic value σ [m] with respect to the average sound absorption coefficient α [m] is large, an appropriate basic value σ [m] (first index value SL1) that matches the auditory tendency described above can be calculated. The special effect is realized.

  The second evaluation unit 344 in FIG. 3 calculates the second index value SL2 corresponding to the M average sound absorption rates α [1] to α [M] calculated by the sound absorption characteristic analysis unit 32. The second index value SL2 is an index for evaluating the liveness of the acoustic space from the viewpoint of the degree of dispersion (the degree of scattering) of the average sound absorption coefficient α [m] for each evaluation band. The second index value SL2 of the first embodiment is any numerical value from 1 (minimum evaluation) to 5 (maximum evaluation) corresponding to each stage (Excellent / Good / Fair / Poor / Bad) that classifies the evaluation results. Set to

  Specifically, the second evaluation unit 344 calculates the dispersion degree D of the M average sound absorption coefficients α [1] to α [M] calculated by the sound absorption characteristic analysis unit 32, and each of the second index values SL2 is calculated. Of the five numerical ranges corresponding to the numerical values (1 to 5), the numerical value (1 to 5) corresponding to the numerical range including the dispersion degree D is calculated as the second index value SL2. Good examples of the spread degree D are standard deviation and variance. The second so that the second index value SL2 becomes larger (numerical value indicating higher evaluation) as the dispersion degree D of the M average sound absorption coefficients α [1] to α [M] is smaller (less scattered). The evaluation unit 344 sets the second index value SL2.

  The index calculation unit 346 of FIG. 3 calculates the evaluation index SL according to the first index value SL1 calculated by the first evaluation unit 342 and the second index value SL2 calculated by the second evaluation unit 344. Specifically, the index calculation unit 346 changes the evaluation index SL from the lowest evaluation 1 to the highest evaluation 5 according to the average (simple average or weighted average) of the first index value SL1 and the second index value SL2 or the added value. Set to any value up to. Accordingly, each average sound absorption coefficient α [m] is close to the predetermined value α0 (the first index value SL1 is large), or the dispersion degree D of M average sound absorption coefficients α [1] to α [M] is small ( As the second index value SL2 is larger, the evaluation index SL is set to a larger numerical value that means higher evaluation.

  The display control unit 28 causes the display device 14 to display the analysis result of the first characteristic analysis unit 261 (evaluation processing unit 34) described above. Specifically, the evaluation screen 81 in FIG. 6, the reverberation characteristic screen 82 in FIG. 7, and the sound absorption characteristic screen 83 in FIG. 8 are selectively displayed on the display device 14 in accordance with an instruction from the user to the input device 15. Is done.

  The evaluation screen 81 in FIG. 6 includes an area 811, an area 812, and an area 813. In the area 811, the evaluation index SL calculated by the evaluation processing unit 34 is displayed. Specifically, among the five-level evaluation results (Excellent / Good / Fair / Poor / Bad), the evaluation results corresponding to the evaluation index SL calculated by the first characteristic analysis unit 261 (index calculation unit 346) (FIG. 6). In the example, “Good”) 814 and a graphic image representing the numerical value of the evaluation index SL (in the illustration of FIG. 6, a graphic image representing the evaluation index SL by the number of stars) 815 are arranged in the region 811. Further, in the area 812, the first index value SL1 calculated by the first evaluation unit 342 is displayed, and in the area 813, the second index value SL2 calculated by the second evaluation unit 344 is displayed. By visually recognizing the evaluation screen 81, the user can quantitatively grasp the suitability of liveness (reverberation) in the acoustic space.

  The reverberation characteristic screen 82 of FIG. 7 is a graph showing reverberation times T [1] to T [M] for each evaluation band with the horizontal axis as the frequency axis. The user can grasp the difference in the reverberation time T [m] for each evaluation band and the general tendency of the reverberation time T [m] in the acoustic space by viewing the reverberation characteristic screen 82. It is.

  The sound absorption characteristic screen 83 of FIG. 8 is a graph showing average sound absorption rates α [1] to α [M] for each evaluation band with the horizontal axis as the frequency axis. By visually recognizing the sound absorption characteristic screen 83, the user grasps the difference in the average sound absorption coefficient α [m] for each evaluation band and the tendency of the approximate average sound absorption coefficient α [m] in the acoustic space. Is possible. The above is the details of the liveness evaluation.

  As described above, in the first embodiment, the evaluation index SL is calculated according to the average sound absorption coefficient α [m] calculated for each evaluation band in the analysis of the impulse response R [n] of the acoustic space. Therefore, for example, when compared with the configuration in which the acoustic characteristics such as the reverberation time T [m] of each evaluation band are presented to the user as the measurement result, it is possible to obtain the acoustic characteristics without requiring specialized knowledge or technical skill regarding the acoustic characteristics. There is an advantage that the user can intuitively and easily grasp the suitability of the acoustic characteristics of the space.

<Flutter echo evaluation>
FIG. 9 is a flowchart of the procedure for evaluating the flutter echo in the acoustic space. 9 is started in response to an instruction from the user to the input device 15 (an instruction to start flutter echo evaluation). The sound emitting device 17 is disposed at a predetermined position in the acoustic space.

  When the process of FIG. 9 is started, the volume adjustment process is executed in the same way as the liveness evaluation illustrated in FIG. 2 (SB1), and the impulse response R is measured at one specific sound receiving point in the acoustic space. (SB2). Specifically, the sound reproduction unit 20 radiates measurement sound from the sound emitting device 17 by supplying the measurement signal X, and the signal acquisition unit 22 observes the observation signal over a predetermined time from the time when the supply of the measurement signal X stops. Y is acquired from the sound collection device 16. The response acquisition unit 24 calculates the impulse response R from the observation signal Y acquired by the signal acquisition unit 22 and stores it in the storage device 12.

  When the measurement of the impulse response R at the sound receiving point is completed by the above-described procedure, the second characteristic analysis unit 262 performs a time axis analysis on components in a specific frequency band (hereinafter referred to as “target band”) of the impulse response R. The envelope ENV is calculated (SB3). The display control unit 28 displays the evaluation result (envelope ENV) by the second characteristic analysis unit 262 on the display device 14 (SB4).

  FIG. 10 is a block diagram of the second characteristic analysis unit 262. As shown in FIG. 10, the second characteristic analysis unit 262 includes a band selection unit 42 and an envelope calculation unit 44. The band selection unit 42 is a band-pass filter that selectively extracts an acoustic component (hereinafter referred to as “target impulse response R0”) in the target band from the impulse response R. The target band is experimentally set so that the influence of the flutter echo is manifested in the envelope ENV. Specifically, a band having a predetermined width including 2 kHz is suitable as the target band. For example, a 1/1 octave band (about 1.4 kHz to about 2.8 kHz) having a center frequency of 2 kHz is set as the target band.

  The envelope calculation unit 44 calculates the envelope ENV on the time axis of the target impulse response R 0 extracted by the band selection unit 42. A known technique can be arbitrarily employed for calculating the envelope ENV, but a method of calculating the envelope ENV by Hilbert transform with respect to the target impulse response R0 is preferable.

  The display control unit 28 causes the display device 14 to display the envelope ENV calculated by the second characteristic analysis unit 262 (envelope calculation unit 44). 11 and 12 are display examples of the envelope ENV. As illustrated in FIGS. 11 and 12, the envelope ENV of the target impulse response R0 is displayed with the horizontal axis as the time axis. When flutter echo is generated in the acoustic space, instantaneous increase in signal intensity periodically occurs in the envelope ENV of the target impulse response R0, as indicated by the broken line in FIG. On the other hand, when no flutter echo is generated in the acoustic space, as shown in FIG. 12, the signal intensity of the envelope ENV decreases substantially linearly with time, and a periodic increase in signal intensity is hardly observed. Therefore, the user can visually grasp the presence or degree of flutter echo in the acoustic space by visually recognizing the envelope ENV displayed on the display device 14.

<Booming evaluation>
FIG. 13 is a flowchart of a procedure for evaluating booming of the acoustic space. The operation of FIG. 13 is started in response to an instruction from the user to the input device 15 (an instruction to start booming evaluation). As illustrated in FIG. 14, the sound collecting device 16 (sound receiving point) and the sound emitting device 17 (sound emitting point) of the acoustic characteristic evaluation device 100 are installed at opposite corners in the acoustic space. In the state, the process of FIG. 13 is executed. Note that the positional relationship between the acoustic characteristic evaluation device 100 and the sound emitting device 17 can be reversed from the illustration of FIG.

  When the process of FIG. 13 is started, the volume adjustment process is executed in the same manner as the liveness evaluation illustrated in FIG. 2 (SC1), and an impulse is received at one sound receiving point in the acoustic space, as in step SB2 of FIG. The response R is measured (SC2). Specifically, the impulse response R is calculated from the observation signal Y over a predetermined time from when the supply of the measurement signal X to the sound emitting device 17 is stopped.

  When the measurement of the impulse response R at the sound receiving point is completed by the above procedure, the third characteristic analyzer 263 calculates the reverberation attenuation characteristic G from the impulse response R in the acoustic space (SC3). The reverberation attenuation characteristic G is a time change (attenuation) of the signal intensity for each frequency component of the impulse response R. Specifically, the third characteristic analysis unit 263 resamples (downsamples) the impulse response R of the predetermined sampling frequency FS1 at the sampling frequency FS2 (FS2 <FS1), and the time waveform after the resampling. Is calculated as a reverberation attenuation characteristic G by the time series of the accumulated phase (phase spectrum) specified by converting the signal into the frequency domain for each frame by known frequency analysis such as fast Fourier transform. The sampling frequency FS1 is, for example, 48 kHz, and the sampling frequency FS2 after re-sampling is, for example, about 1 kHz. Since booming is a problem in the low sound range as described above, it is possible to calculate the reverberation attenuation characteristic G for the low frequency side band (for example, 50 Hz to 500 Hz) of the impulse response R.

  When the above processing is completed, the display control unit 28 causes the display device 14 to display the evaluation result (reverberation attenuation characteristic G) by the third characteristic analysis unit 263 (SC4). 15 and 16 are display examples of the reverberation attenuation characteristic G. FIG. As illustrated in FIG. 15 and FIG. 16, the gradation and color of each point on the coordinate plane where the time axis (horizontal axis) and the frequency axis (vertical axis) are installed are set according to the signal strength of the reverberation attenuation characteristic G. By setting, the distribution of the reverberation attenuation characteristic G is displayed. FIG. 15 is a display example of the reverberation attenuation characteristic G when booming occurs in the acoustic space, and FIG. 16 shows that booming is suppressed by installing a sound field control panel (articulation panel) in the acoustic space. It is an example of a display of the reverberation attenuation characteristic G in the case. When booming (standing wave) is generated in the acoustic space, as shown in FIG. 15, the local frequency component continues for a long time, and the band between each frequency component also increases or decreases in time. Repeated acoustic components are observed. That is, the standing wave in the acoustic space is visualized. On the other hand, when booming does not occur in the acoustic space, as shown in FIG. 16, the continuation of local frequency components over a long period of time is hardly observed. Accordingly, the user can visually grasp the presence or degree of booming in the acoustic space by visually recognizing the reverberation attenuation characteristic G displayed on the display device 14.

Second Embodiment
A second embodiment of the present invention will be described. In each of the embodiments exemplified below, elements having the same functions and functions as those of the first embodiment are referred to in the description of the first embodiment, and detailed descriptions thereof are appropriately omitted.

  In the first embodiment, the impulse integration method is used to calculate the reverberation time T [m] for each evaluation band. In the impulse integration method, the time length of the section selected as the integration range (hereinafter referred to as “extraction section”) in the band impulse response RB [n, m] of each evaluation band is determined by whether the reverberation curve C [n, m] is appropriate. There is a tendency to influence. For example, assume that the reverberation curve (ideal reverberation curve) C [n, m] in FIG. 4 is calculated when the extraction interval is set to an appropriate time length. When the time length of the extraction section is longer than the optimum value, the signal strength on the tail side is emphasized as shown by the reverberation curve CL [n, m] shown by the broken line in FIG. 4, so that the band impulse response RB [ The influence of the electrical or acoustic noise component superimposed on n, m] relatively increases, and as a result, the accuracy of specifying the reverberation time T [m] decreases. On the other hand, when the time length of the extraction section is shorter than the optimum value, the signal strength on the tail side is excessively suppressed as shown by the reverberation curve CS [n, m] shown by the broken line in FIG. Accurate specification of the time T [m] is hindered. Considering the above circumstances, in the second embodiment of the present invention, the extraction interval of the band impulse response RB [n, m] is set so that the reverberation time T [m] can be specified with high accuracy.

  The acoustic characteristic evaluation apparatus 100 of the second embodiment has a configuration in which the reverberation time analysis unit 324 of the first embodiment is replaced with a reverberation time analysis unit 324A of FIG. As shown in FIG. 17, the reverberation time analysis unit 324A includes a first analysis unit 52, a reverberation time provisional unit 54, a second analysis unit 56, and a reverberation time calculation unit 58.

  As shown in FIG. 18, the first analysis unit 52 calculates a reverberation curve C1 [n, m] corresponding to the extraction section S1 in the band impulse response RB [n, m] processed by the band dividing unit 322. . Specifically, the first analysis unit 52 sets an extraction interval S1 extending from the start point of the band impulse response RB [n, m] to a predetermined length, and uses a reverberation curve C1 by an impulse integration method with the extraction interval S1 as an integration range. Calculate [n, m]. The time length of the extraction section S1 is set experimentally or statistically so as to be longer than the optimum value of the integration range described with reference to FIG.

  As shown in FIG. 18, the reverberation time provisional unit 54 determines a provisional reverberation time (hereinafter referred to as “provisional reverberation time”) T0 [n, m from the reverberation curve C1 [n, m] calculated by the first analysis unit 52. ] Is set. Specifically, the reverberation time provisional unit 54 calculates the provisional reverberation time T0 [n, m] according to the time length τ in which the signal intensity attenuates by a predetermined amount in the reverberation curve C1 [n, m]. As described above, since the extraction section S1 is set to a longer time length than the optimum value, the signal strength on the tail side of the reverberation curve C1 [n, m] is ideal as shown by a broken line in FIG. A reverberation curve (hereinafter referred to as “target reverberation curve”) C0. However, in the reverberation curve C1 [n, m], the section immediately after t0 when the signal intensity starts attenuation (hereinafter referred to as “attenuation start point”) (initial attenuation section) approximates the target reverberation curve C0. There is a tendency. Considering the above tendency, in the second embodiment, the time length τ is specified according to the section immediately after the attenuation start point t0 in the reverberation curve C1 [n, m].

  Specifically, the signal strength is increased from the time t1 when the signal strength is decreased by a predetermined amount Δ1 (for example, 5 dB) as compared with the initial value m0 at the attenuation start point t0 under the reverberation curve C1 [n, m]. The signal strength is attenuated by a predetermined amount (Δ2−Δ1) immediately after the time t2 when the predetermined amount Δ2 (for example, 15 dB) lower than the initial value m0 is decreased by a predetermined amount Δ2 (for example, 15 dB). Time) is specified as the time length τ. Since the reverberation time is defined as a time length in which the signal intensity is attenuated by 60 dB, the reverberation time provisional unit 54 sets the time length τ in which the signal intensity is attenuated by a predetermined amount (Δ2−Δ1) to {60 / (Δ2−Δ1). )} Is calculated as the provisional reverberation time T0 [n, m].

  As shown in FIG. 18, the second analysis unit 56 in FIG. 17 includes the provisional reverberation time T0 [m] set by the reverberation time provisional unit 54 in the band impulse response RB [n, m] processed by the band division unit 322. ], The reverberation curve C2 [n, m] corresponding to the extraction section S2 is calculated. Specifically, the second analysis unit 56 sets an extraction interval S2 that extends from the start point of the band impulse response RB [n, m] to a predetermined length, and uses a reverberation curve C2 by an impulse integration method with the extraction interval S2 as an integration range. Calculate [n, m].

  The time length of the extraction section S2 is variably set according to the provisional reverberation time T0 [m] set by the reverberation time provisional unit 54. Specifically, the extraction interval S2 is set to a time length (κT0 [n, m]) obtained by multiplying the provisional reverberation time T0 [n, m] by the coefficient κ. The coefficient κ is set to a numerical value greater than 1, for example, preferably set to 2. As understood from the above description, the extraction section S1 set by the first analysis unit 52 and the extraction section S2 set by the second analysis unit 56 may have different time lengths. Specifically, there is a high possibility that the extraction section S2 is set to a shorter section than the extraction section S1.

  The reverberation time calculation unit 58 in FIG. 17 calculates a definite reverberation time T [m] from the reverberation curve C2 [n, m] calculated by the second analysis unit 56. Specifically, the reverberation time calculation unit 58 generates reverberation curves C2 [1, m] to C2 [N, m] of the band impulse responses RB [m] of N reception points corresponding to the mth evaluation band. ], The reverberation time T [m] of the mth evaluation band is calculated from the reverberation curve CA [m]. For example, the time length during which the signal intensity decreases by 60 dB under the reverberation curve CA [m] obtained by averaging N reverberation curves C2 [1, m] to C2 [N, m] (or, for example, the signal intensity is ( 60 / β) β times the length of time reduced by dB) is calculated as the reverberation time T [m].

  The reverberation time calculation unit 58 calculates the reverberation time T [m] (T [1] to T [M]) for each evaluation band by executing the above process for each of the M evaluation bands. In the above example, the reverberation time T [m] is calculated from the reverberation curves CA [m] corresponding to the N reverberation curves C2 [1, m] to C2 [N, m]. A numerical value corresponding to N reverberation times specified from each reverberation curve C2 [n, m] (C2 [1, m] to C2 [N, m]) corresponding to the band (for example, an average of N reverberation times) Value) as the reverberation time T [m]. Processing using the reverberation time T [m] calculated by the reverberation time calculation unit 58 (calculation of the average sound absorption coefficient α [m] and display of the evaluation result) is the same as in the first embodiment.

  In the second embodiment, the same effect as in the first embodiment is realized. In the second embodiment, the provisional reverberation time T0 [n, m] is specified from the reverberation curve C1 [n, m] corresponding to the extraction section S1 in the band impulse response RB [n, m], and the band impulse response is determined. The reverberation time T [m] is calculated from the reverberation curve C2 [n, m] corresponding to the extraction section S2 having a time length corresponding to the provisional reverberation time T0 [n, m] in RB [n, m]. Therefore, there is an advantage that the reverberation time T [m] can be specified with high accuracy by reducing the influence of the length of the extraction interval set in the band impulse response RB [n]. For example, when the extraction interval S1 is set to a sufficiently long time length as compared with the optimum value as described above, the provisional reverberation time T0 [n, m] specified from the reverberation curve C1 [n, m] is the extraction interval. Compared to S1, the time length is short (that is, the time length is close to the optimum value). Therefore, the reverberation curve C2 [n, m] specified by the second analysis unit 56 is closer to the target reverberation curve C0 than the reverberation curve C1 [n, m], and as a result, the reverberation time T [m] is highly accurate. Specified.

  In the second embodiment, the time length during which the signal intensity attenuates by a predetermined amount (10 dB below 60 dB, which is the original reference for the reverberation time) immediately after the attenuation start point t0 in the reverberation curve C1 [n, m]. The provisional reverberation time T0 [n, m] is calculated according to τ. Accordingly, the provisional reverberation time T0 [n, m] is calculated according to the signal strength of the end of the reverberation curve C1 [n, m] (that is, the section where the reverberation curve C1 deviates from the target reverberation curve C0). Compared with a configuration in which the signal intensity decays by 60 dB directly from the reverberation curve C1 [n, m] as the provisional reverberation time T0 [n, m], the target of the reverberation curve C1 [n, m] A provisional reverberation time T0 [n, m] corresponding to a section approximating the reverberation curve C0 is calculated. Therefore, the effect that the reverberation time T [m] can be calculated with high accuracy is particularly remarkable.

<Modification>
Each of the above forms can be variously modified. Specific modifications are exemplified below. Two or more aspects arbitrarily selected from the following examples can be appropriately combined.

(1) In each of the above-described embodiments, a single measurement result in the acoustic space is displayed. However, a plurality of measurement results can be displayed in a comparative manner. For example, the measurement of the acoustic characteristics described in the above embodiments is repeated a plurality of times, and the evaluation index SL (SL1, SL2) calculated by the first characteristic analysis unit 261 and the envelope ENV calculated by the second characteristic analysis unit 262 are obtained. In addition, a configuration in which the display control unit 28 displays the reverberation attenuation characteristics G calculated by the third characteristic analysis unit 263 on the display device 14 at a plurality of different points in time is preferable. Examples of the configuration for displaying a plurality of acoustic characteristics in comparison may include a configuration for displaying the acoustic characteristics in parallel in different areas and a configuration for displaying the acoustic characteristics in duplicate. According to the above configuration, for example, the user can intuitively understand the effect of the acoustic characteristic adjustment by displaying the measurement results before and after the acoustic characteristic adjustment of the acoustic space (for example, installation of the sound field control panel) in a comparative manner. There is an advantage of being able to grasp.

(2) The method by which the first characteristic analysis unit 261 calculates the evaluation index SL is appropriately changed. For example, a configuration in which the first evaluation unit 342 of the first characteristic analysis unit 261 calculates the basic value σ [m] by the following equation (3) is also preferable.

As understood from Equation (3), the larger one of the distance ΔL and the distance ΔH is adopted as the basic value σ [m]. The distance ΔL is calculated by the following formula (4A), and the distance ΔH is calculated by the following formula (4B).

  FIG. 19 is a graph showing the relationship between the average sound absorption coefficient α [m] and the basic value σ [m]. As understood from the equation (4A) and FIG. 19, in the range QL where the average sound absorption coefficient α [m] is lower than the predetermined value αL, the basis is set according to the logarithmic distance between the average sound absorption coefficient α [m] and the predetermined value αL. The value σ [m] is calculated, the basic value σ [m] is set to 0 in the range QM from the predetermined value αL to the predetermined value αH, and in the range QH where the average sound absorption coefficient α [m] exceeds the predetermined value αH, The basic value σ [m] is calculated according to the logarithmic distance between the average sound absorption coefficient α [m] and the predetermined value αH. A range QM from the predetermined value αL to the predetermined value αH means a suitable range (allowable range) of the average sound absorption coefficient α [m]. Specifically, the predetermined value αL and the predetermined value αH are set to numerical values of about 0.15 to 0.35 depending on, for example, the use of the acoustic space.

  Even in the above configuration, the same effect as in the first embodiment is realized. Further, when the average sound absorption coefficient α [m] is a numerical value within the range QM, the basic value σ [m] is set to 0, and therefore, out of the M average sound absorption coefficients α [1] to α [M]. There is also an advantage that the first index value SL1 can be set to an appropriately large numerical value (a numerical value indicating high evaluation) when the number of average sound absorption coefficients α [m] within the range QM is large.

(3) In the liveness evaluation of the first embodiment, the impulse response R [n] (R [1] to R [N]) is measured at each of a plurality (N) of sound receiving points. It is also possible to measure the impulse response R for only the sound points. The reverberation time analysis unit 324 reverberates from the reverberation curve C [m] corresponding to each of the M band impulse responses RB [1] to RB [M] obtained by dividing one impulse response R for each evaluation band. [m] (T [1] to T [M]) is calculated.

  Similarly in the second embodiment, it is possible to measure the impulse response R for only one sound receiving point. In the reverberation time analysis unit 324A, for each of the M band impulse responses RB [1] to RB [M] obtained by dividing one impulse response R for each evaluation band, a reverberation curve C1 [m ] And the provisional reverberation time T0 [m] are calculated, and the reverberation time T [m] is calculated from the reverberation curve C2 [m] corresponding to the extraction section S2 corresponding to the provisional reverberation time T0 [m]. The

(4) In the first embodiment, the acoustic characteristic evaluation apparatus 100 that can perform liveness evaluation of the acoustic space, flutter echo evaluation, and booming evaluation has been illustrated. However, one or two of the three types of evaluations illustrated above may be used. An acoustic characteristic evaluation apparatus 100 that performs types of evaluation can also be realized. That is, the acoustic characteristic evaluation apparatus 100 is included as an apparatus including at least one of the first characteristic analysis unit 261, the second characteristic analysis unit 262, and the third characteristic analysis unit 263.

(5) In the second embodiment, the reverberation time T [m] calculated by the reverberation time analysis unit 324A is used for liveness evaluation. However, the second embodiment is also used as a reverberation time analysis apparatus for calculating the reverberation time of the acoustic space. Can be realized. That is, liveness evaluation is not an essential requirement for the configuration of the second embodiment. In the liveness evaluation exemplified in the above embodiments, the reverberation time T [m] is calculated for each evaluation band in order to calculate the average sound absorption coefficient α [m] of each evaluation band. However, attention should be paid only to the calculation of the reverberation time. For example, processing for each evaluation band is not essential. In other words, the reverberation time analysis device (reverberation time analysis unit 324A) includes a first analysis unit 52 that specifies the reverberation curve C1 corresponding to the extraction section S1 of the impulse response R of the acoustic space (whether or not band division is performed), and the reverberation. A reverberation time provisional unit 54 for setting the provisional reverberation time T0 according to the time when the signal intensity attenuates by a predetermined amount in the curve C1, and an extraction section over a time length (for example, κT0) corresponding to the provisional reverberation time T0 in the impulse response R. It is expressed as an apparatus including a second analysis unit 56 that specifies a reverberation curve C2 corresponding to S2, and a reverberation time calculation unit 58 that calculates a reverberation time T from the reverberation curve C2.

(6) The content of flutter echo evaluation by the second characteristic analysis unit 262 is not limited to the above examples. For example, as shown in FIG. 20, a configuration in which a reverberation curve specifying unit 46 and an evaluation processing unit 48 are added to the second characteristic analysis unit 262 of each of the above-described embodiments may be employed. The reverberation curve specifying unit 46 specifies the reverberation curve C from the target impulse response R 0 after processing by the band selecting unit 42. For example, the impulse integration method is preferably used to specify the reverberation curve C. Similarly to the second embodiment, the provisional reverberation time T0 is calculated from the reverberation curve C1 corresponding to the extraction section S1 of the impulse response R in the acoustic space, and the time length corresponding to the provisional reverberation time T0 of the impulse response R is calculated. It is also possible to calculate the reverberation curve C (C2) from the extraction section S2. A configuration in which the reverberation curve specifying unit 46 calculates the reverberation curve C from the impulse response R before processing by the band selection unit 42 may also be employed.

  While the influence of the flutter echo becomes obvious in the envelope ENV calculated by the envelope calculation section 44, the influence of the flutter echo is reduced in the reverberation curve C of the impulse response R. Therefore, when no flutter echo is generated in the acoustic space, the envelope ENV and the reverberation curve C are approximated, and the difference between the envelope ENV and the reverberation curve C is increased as the flutter echo in the acoustic space becomes more prominent. There is a tendency to increase. In consideration of the above tendency, the evaluation processing unit 48 of FIG. 20 performs an acoustic space operation in accordance with the difference between the envelope ENV calculated by the envelope calculation unit 44 and the reverberation curve C specified by the reverberation curve specifying unit 46. An evaluation index SF that is a measure of the degree of occurrence of flutter echo is calculated.

  For example, the evaluation processing unit 48 calculates the difference value | ENV−C | of the signal strength at each of a plurality of time points on the time axis between the envelope ENV and the reverberation curve C, and the difference value | corresponding to each time point | The evaluation index SF is calculated from ENV-C |. For example, an accumulated value or an average value of difference values | ENV-C | over a plurality of time points is calculated as the evaluation index SF. Therefore, the evaluation index SF becomes a larger numerical value as the flutter echo in the acoustic space becomes more conspicuous (the difference between the envelope ENV and the reverberation curve C is more conspicuous). The display control unit 28 causes the display device 14 to display the evaluation result (for example, the evaluation index SF) by the evaluation processing unit 48. A configuration for displaying the five-stage evaluation according to the evaluation index SF and a configuration for displaying the evaluation result by the evaluation processing unit 48 together with the envelope ENV illustrated in FIGS. 11 and 12 may be employed. According to the above structure, there exists an advantage that a user can grasp | ascertain the generation | occurrence | production degree of the flutter echo in acoustic space quantitatively.

(7) In the above-described booming evaluation of each form, the reverberation attenuation characteristic G is calculated according to the accumulated phase after the resampling of the impulse response R. However, the calculation method of the reverberation attenuation characteristic G is not limited to the above examples. For example, the reverberation curve C can be calculated for each frequency component of the impulse response R specified by the response acquisition unit 24, and the reverberation attenuation characteristic G can be calculated by arranging the reverberation curve C on the frequency axis. For example, the impulse integration method is preferably used to specify the reverberation curve C. Similarly to the second embodiment, the provisional reverberation time T0 is calculated from the reverberation curve C1 corresponding to the extraction section S1 of the impulse response R in the acoustic space, and the time length corresponding to the provisional reverberation time T0 of the impulse response R is calculated. It is also possible to calculate the reverberation curve C (C2) from the extraction section S2. Note that, as illustrated in the above-described embodiments, according to the configuration in which the reverberation attenuation characteristic G is calculated from the accumulated phase after resampling with respect to the impulse response R, the reverberation attenuation characteristic is determined from the reverberation curve C for each frequency component of the impulse response R. There is an advantage that the processing load (processing amount and processing time) is greatly reduced as compared with the configuration for calculating G.

(8) In each of the above-described embodiments, the acoustic characteristic evaluation device 100 is realized by a portable information processing device such as a mobile phone or a smartphone, but a server that communicates with a terminal device via a communication network such as a mobile communication network or the Internet. It is also possible to realize the acoustic characteristic evaluation apparatus 100 with an apparatus (for example, a web server). That is, the acoustic characteristic evaluation device 100 uses the response acquisition unit 24 that acquires the impulse response R corresponding to the observation signal Y collected by the terminal device in the acoustic space, and the impulse response R acquired by the response acquisition unit 24. And an analysis processing unit 26 that analyzes the acoustic characteristics of the acoustic space. The response acquisition unit 24 acquires, from the terminal device, an element that specifies the impulse response R from the observation signal Y transmitted from the terminal device to the acoustic characteristic evaluation device 100 or the impulse response R that is specified from the observation signal Y by the terminal device. It is an element to do. The analysis result by the analysis processing unit 26 is transmitted from the acoustic characteristic evaluation device 100 to the terminal device and notified (for example, displayed) to the user of the terminal device. Note that it is also possible for the acoustic characteristic evaluation device 100 (server device) to execute only a part of the processing of the analysis processing unit 26 exemplified in the above-described embodiments (for example, calculation of the reverberation time exemplified in the second embodiment). is there.

DESCRIPTION OF SYMBOLS 100 ... Acoustic-characteristic evaluation apparatus, 10 ... Arithmetic processing apparatus, 12 ... Memory | storage device, 14 ... Display apparatus, 15 ... Input device, 16 ... Sound collection apparatus, 17 ... Sound emission apparatus, 20 ... Sound reproduction unit, 22 ... signal acquisition unit, 24 ... response acquisition unit, 26 ... analysis processing unit, 261 ... first characteristic analysis unit, 262 ... second characteristic analysis unit, 263 ... third characteristic analysis , 28 ... Display control unit, 32 ... Sound absorption characteristic analysis unit, 322 ... Band division unit, 324 ... Reverberation time analysis unit, 326 ... Sound absorption rate calculation unit, 34 ... Evaluation processing unit, 342 ... First evaluation unit, 344... Second evaluation unit, 346... Index calculation unit, 42... Band selection unit, 44... Envelope calculation unit, 46. 52 …… First analysis section, 54 …… Reverberation time provisional section, 56 …… Second analysis section, 58 …… Reverberation time Tough.

Claims (3)

First analysis means for specifying a first reverberation curve corresponding to a first extraction section of an impulse response of an acoustic space;
Reverberation time provisional means for calculating provisional reverberation time according to the time when the signal intensity attenuates by a predetermined amount in the first reverberation curve;
Second analysis means for specifying a second reverberation curve corresponding to a second extraction interval over a time length corresponding to the provisional reverberation time in the impulse response;
A reverberation time analysis device comprising: reverberation time calculation means for calculating reverberation time from the second reverberation curve. The reverberation time analysis apparatus according to claim 1, wherein the second analysis means sets the second extraction section over a time length obtained by multiplying the provisional reverberation time by a predetermined value. 3. The reverberation time according to claim 1, wherein the reverberation time provisional unit sets the provisional reverberation time according to a time length in which the signal intensity attenuates by a predetermined amount immediately after the attenuation start point in the first reverberation curve. Analysis device.