JP2019032242A - Attenuation time analytic method, device, and program - Google Patents

Abstract

To determine an attenuation time of a sound having a specific frequency in a target space without using a band pass filter corresponding to the frequency.SOLUTION: An attenuation time analysis device 100 determines an energy attenuation curve corresponding to a specific frequency on the basis of the result of a movement Fourier transform of an impulse response of a sound in a target space. Then, the attenuation time analysis device 100 determines the attenuation time of the sound having the specific frequency in the target space in response to the gradient of a regression line obtained from the energy attenuation curve.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a decay time analysis method, apparatus, and program.

  Conventionally, an automatic reverberation time measuring device that measures reverberation time is known (see, for example, Patent Document 1). In the technique described in Patent Document 1, first, a time average S / N that represents an S / N every moment is obtained from the direct sound arrival time in the impulse response. Then, the time zone from the time when the direct sound arrives in the impulse response to the time when the time average S / N first becomes 0 dB is detected, and based on the time zone until the time when the time average S / N first becomes 0 dB, The observation section is determined. Further, the slope of the least square approximation line of the acoustic energy attenuation curve in the determined observation section is calculated, and the reverberation time is calculated from the obtained slope.

JP-A-8-247837

  However, when the attenuation time (or reverberation time) of a specific frequency is obtained using the technique described in Patent Document 1, it is necessary to prepare a band filter corresponding to the specific frequency. A signal corresponding to a specific frequency is extracted by the bandpass filter. Further, in order to obtain attenuation times of a plurality of different frequencies, it is necessary to prepare a band filter for each frequency in order to extract a signal corresponding to a specific frequency.

  In view of the above facts, the present invention has an object to obtain the decay time of sound of a specific frequency in the target space without using a bandpass filter according to the frequency.

  In order to achieve the above object, the decay time analysis method of the present invention obtains an energy decay curve corresponding to a specific frequency from the result of the moving Fourier transform of the sound impulse response in the target space, and obtains it from the energy decay curve. The decay time of the sound of the specific frequency in the target space is obtained according to the slope of the regression line. Thereby, the attenuation time of the sound of the specific frequency in object space can be calculated | required, without using the band filter according to the frequency.

  The attenuation time analysis method of the present invention calculates an attenuation level curve of the energy attenuation curve, and based on a predetermined analysis interval corresponding to a level range of attenuation of the smoothed waveform of the attenuation level curve, the attenuation time curve in the analysis interval The slope of the regression line of the smoothed waveform can be calculated, and the decay time of the sound of the specific frequency in the target space can be obtained according to the calculated slope of the regression line of the smoothed waveform. Thereby, even if the time change of the attenuation level curve obtained from the energy attenuation curve is abrupt, the attenuation time of the sound of a specific frequency in the target space can be obtained with high accuracy.

  The decay time analysis method of the present invention obtains an energy decay curve corresponding to the predetermined frequency band by obtaining an average of the energy decay curves in the predetermined frequency band from the result of the moving Fourier transform, and obtains the energy decay. The decay time of the sound of the predetermined frequency band in the target space can be obtained according to the slope of the regression line obtained from the curve. Thereby, the attenuation time of the sound in a specific frequency band in the target space can be obtained without using a band filter corresponding to the frequency band.

  The decay time analysis method of the present invention obtains an average of the energy decay curves in a plurality of different frequency bands or in a plurality of different frequencies from the result of the moving Fourier transform, so that the plurality of different frequency bands or the plurality of different ones. An energy attenuation curve corresponding to the frequency is obtained, and the sound attenuation times of the plurality of different frequency bands or the plurality of different frequencies in the target space are obtained according to the slope of the regression line obtained from the energy attenuation curve. Can be. Thereby, without using a plurality of different frequency bands or a plurality of different frequency band filters, it is possible to obtain an attenuation time of a sound in which sounds of a plurality of different frequency bands or a plurality of different frequencies are superimposed in the target space. it can.

  The decay time analysis apparatus of the present invention is calculated by an energy curve calculation unit for obtaining an energy decay curve corresponding to a specific frequency from the result of the moving Fourier transform of the impulse response of the sound in the target space, and the energy curve calculation unit. An attenuation time calculation unit that calculates an attenuation time of the sound of the specific frequency in the target space according to the slope of the regression line obtained from the energy attenuation curve.

  The program of the present invention is calculated by an energy curve calculation unit that obtains an energy decay curve corresponding to a specific frequency from the result of a moving Fourier transform of a sound impulse response in a target space, and the energy curve calculation unit. It is a program for functioning as an attenuation time calculation unit for obtaining the attenuation time of the sound of the specific frequency in the target space according to the slope of the regression line obtained from the energy attenuation curve.

  According to the present invention, there is an effect that it is possible to obtain the decay time of a sound having a specific frequency in the target space without using a band filter corresponding to the frequency.

It is a block diagram which shows schematic structure of the decay time analysis system which concerns on embodiment of this invention. It is a figure which shows an example of the energy decay curve (level (dB) display) of each frequency obtained by the moving Fourier transform of an impulse response. It is explanatory drawing for demonstrating the smoothing waveform and regression line of the attenuation level curve of a specific frequency. It is a figure which shows an example of a decay time analysis processing routine. It is explanatory drawing for demonstrating the superimposition of the attenuation level curve of a several different frequency. It is a figure which shows an example of a simulation result.

  Hereinafter, embodiments of the present invention will be described in detail.

[First Embodiment]
<Configuration of decay time analysis system>

  FIG. 1 is a block diagram showing an example of the configuration of an attenuation time analysis system 100 according to the first embodiment of the present invention. The decay time analysis system 100 calculates the decay time of the sound in the target space. The sound in the target space is a sound generated in the target space, for example, a sound generated by a human voice or a musical instrument.

The sound attenuation time in the target space is a physical quantity that defines the attenuation characteristics. In addition, the decay time of sound not only represents the transient state characteristic of attenuation, but also relates to the magnitude of steady state energy because it follows the process of decaying steady state energy after sound is emitted. . An example of a sound field generated by sound generated in the target space is a room sound field.
Note that the phenomenon of attenuation is a phenomenon common to all wave fields including, for example, a vibration field in a finite object, and the attenuation time can be defined similarly to the attenuation time of a sound field. .

  In order to grasp the nature of the sound field, it is necessary to accurately calculate the decay time. In general, since the attenuation characteristic depends on the frequency, the attenuation time for each band can be obtained by dividing the response signal of the sound field into each frequency band. Specifically, a method using a band noise and a method using an impulse can be cited as methods for obtaining the decay time.

  In the method using band noise, first, a band-limited noise signal is output from a source (for example, a speaker or an exciter) by applying a band filter to the noise signal, and the sound field is in a steady state. , Stop the band-limited noise signal. At this time, a response signal to the noise signal is recorded by a sensor (for example, a microphone or a vibration pickup), an energy decay curve obtained from the response signal is observed, and an decay time is obtained from the slope of the energy decay curve.

In the method using an impulse, first, an impulse signal is output from the source, and the impulse response between the source and the sensor is measured. Next, the impulse response is passed through a band filter to obtain a band impulse h _b (t). Further, an energy decay curve G _b (t) is obtained by performing Schroeder integration represented by the following equation (1) on the impulse response h _b (t). Then, the decay time is obtained from the slope of the energy decay curve G _b (t).

  As described above, when the attenuation time is obtained using the method using band noise and the method using impulse, it is necessary to use a band filter. However, since the narrower the band, the more difficult it is to realize an appropriate characteristic, the cutoff characteristic often deteriorates. Further, since the band filter has a long transient, it takes time to convolve with the original signal.

  Therefore, in this embodiment, the decay time in a desired band is estimated from the energy decay curve of the sound field excited in the target space with a single sine wave. Specifically, in the present embodiment, the moving Fourier transform of the sound impulse response in the target space is calculated. Then, an energy decay curve corresponding to a specific frequency is obtained according to the result of the moving Fourier transform, and an attenuation time of a sound having a specific frequency in the target space is obtained from the energy decay curve. This will be specifically described below.

  As shown in FIG. 1, the decay time analysis system 100 can be functionally represented by a configuration including a reception unit 10, an decay time analysis device 20, and an output unit 40.

  The accepting unit 10 accepts information input from the outside. The receiving unit 10 receives an impulse response h (t) of sound in the target space, which is input from an input device (not shown). Moreover, the reception part 10 acquires the specific frequency input from the input device. The impulse response h (t) of the sound in the target space is measured in advance by, for example, outputting an impulse signal from a speaker installed in the target space and measuring with a microphone.

  The decay time analyzer 20 includes a CPU (Central Processing Unit), a ROM (Read Only Memory) that stores programs for realizing each processing routine, a RAM (Random Access Memory) that temporarily stores data, and storage means. This is realized by a computer including a memory, a network interface, and the like. As shown in FIG. 1, the decay time analyzer 20 functionally includes a response result acquisition unit 22, a moving Fourier transform analysis unit 24, an analysis result storage unit 26, a frequency setting unit 27, an energy curve. The generation unit 28, the attenuation level curve generation unit 30, the smoothing unit 32, the analysis interval setting unit 34, the regression calculation unit 36, and the attenuation time calculation unit 38 can be represented.

  The response result acquisition unit 22 acquires the impulse response h (t) of the sound in the target space received by the reception unit 10.

  When the impulse response h (t) of the sound in the target space is acquired, the energy decay curve G (f, t) and the impulse response h (t) when the target space is excited with a single sine wave of the frequency f The following equation (2) holds (for example, reference (Yuo Yamada, “A consideration on attenuation frequency characteristics and attenuation density of room sound field”, Acoustical Society of Japan Proceedings 2016 Autumn) See).

The relationship of the above equation (2) is that a sound field excited by a single sine wave of frequency f ₁ is obtained by obtaining a moving Fourier transform of the impulse response h (t) and then extracting a component of a specific frequency f _1. It is shown that an energy decay curve is obtained. The moving Fourier transform is to make the integration start point t variable and perform Fourier transform for each t. The integral part in the above equation (2) corresponds to the Fourier transform for t.

Therefore, in this embodiment, the moving Fourier transform of the impulse response h (t) is calculated, and the energy decay curve G (f, t) with the time t as a variable is acquired in advance for each frequency f. Next, an energy attenuation curve of a specific frequency f ₁ that is an object of calculation of the attenuation time is obtained from the energy attenuation curve G (f, t) of each frequency f. Then, the decay time of the specific frequency f ₁ is calculated according to the energy decay curve G (f ₁ , t). Accordingly, without using a band filter, it is possible to obtain a decay time of a particular frequency f _1.

  Therefore, the moving Fourier transform analysis unit 24 performs a moving Fourier transform on the impulse response h (t) acquired by the response result acquisition unit 22. Then, the moving Fourier transform analysis unit 24 stores the moving Fourier transform result for the impulse response h (t) in the analysis result storage unit 26.

  The analysis result storage unit 26 stores the result of the moving Fourier transform obtained by the moving Fourier transform analysis unit 24. FIG. 2 shows an example of the result of the moving Fourier transform for the impulse response. As shown in FIG. 2, an energy decay curve with time as a variable is obtained for each frequency by the moving Fourier transform on the impulse response. Note that the energy decay curve shown in FIG. 2 is displayed in level (dB).

The frequency setting unit 27 sets a specific frequency f ₁ according to the information received by the receiving unit 10. For example, information regarding the specific frequency f ₁ is input to the reception unit 10 by the operator of the decay time analysis apparatus 20.

The energy curve generation unit 28 obtains an energy decay curve G (f ₁ , t) corresponding to the specific frequency f ₁ set by the frequency setting unit 27 from the result of the moving Fourier transform stored in the analysis result storage unit 26. To do.

Attenuation level curve generator 30, the energy attenuation acquired by the energy curve generating unit 28 curve G (f _{1, t)} was logarithmic representation, the attenuation level curve of the energy decay curve G (f _{1, t)} calculated To do. When the energy decay curve G (f ₁ , t) obtained by the above equation (2) is logarithmically expressed ( ₁₀ log ₁₀ G (f, t)), an attenuation level curve GL (f ₁ , t) is obtained. An example of the attenuation level curve is shown in FIG. As shown in FIG. 3, the attenuation level curve GL has a shape that linearly descends.

Smoothing unit 32 performs the time averaging processing is applied is calculated by the attenuation level curve generating unit 30 attenuation level curve GL _(f 1, t), the smoothed waveform of the attenuation level curve GL _(f 1, t) obtain. Specifically, the smoothing unit 32 sets a predetermined time window and averages the values in the time window with respect to the waveform representing the attenuation level curve GL (f ₁ , t), so that the attenuation level curve GL is obtained. A smoothed waveform of (f ₁ , t) is obtained. As shown in FIG. 3, since the attenuation level curve GL is accompanied by vibration (fluctuation), the smoothed waveform M subjected to the time averaging process by the smoothing unit 32 is used in the subsequent processes.

  The analysis interval setting unit 34 sets an analysis interval for the smoothed waveform of the attenuation level curve obtained by the smoothing unit 32. Specifically, the analysis interval setting unit 34 sets a predetermined analysis interval representing a time range corresponding to the attenuation level range of the smoothed waveform of the attenuation level curve. The analysis section is set according to a preset level range (for example, -5 dB to -35 dB, etc.). In the example shown in FIG. 3, an analysis interval in which the value of the smoothed waveform M of the attenuation level curve GL is −10 dB to −35 dB is set.

  The regression calculation unit 36 calculates a regression line of the smoothed waveform in the analysis section based on the smoothed waveform in the analysis section set by the analysis section setting unit 34. For example, the regression calculation unit 36 obtains the gradient of the smoothed waveform in the analysis interval by the least square method. In the example shown in FIG. 3, a regression line R is obtained for the analysis interval.

  The decay time calculation unit 38 calculates the decay time of a sound having a specific frequency in the target space based on the slope of the regression line calculated by the regression calculation unit 36. By obtaining the gradient of the regression line R calculated by the regression calculation unit 36, the decay time of the sound in the target space when the target space is excited by a single sine wave having a specific frequency is obtained.

The output unit 40 outputs the decay time of the sound of the specific frequency f ₁ in the target space, calculated by the decay time calculation unit 38 as a result. The output unit 40 is, for example, is realized by a display or the like, is displayed decay time of a particular frequency f ₁ of the sound.

<Operation of decay time analysis system>

  Next, the operation of the decay time analysis system 100 will be described with reference to FIG. When the decay time analysis system 100 is activated and the input of the impulse response h (t) in the target space is received by the receiving unit 10, the decay time analysis device 20 executes the decay time analysis processing routine shown in FIG.

  In step S <b> 100, the response result acquisition unit 22 acquires the impulse response h (t) of the sound in the target space received by the reception unit 10.

  In step S102, the moving Fourier transform analysis unit 24 performs a moving Fourier transform on the impulse response h (t) acquired in step S100. Then, the moving Fourier transform analysis unit 24 stores the moving Fourier transform result for the impulse response h (t) in the analysis result storage unit 26.

In step S <b> 103, the frequency setting unit 27 sets a specific frequency f ₁ according to the information received by the receiving unit 10.

In step S104, the energy curve generator 28, from the mobile Fourier transform result stored into the analysis result storage unit 26 in step S102, corresponds to a specific frequency f ₁ set in step S103 the energy decay curve G ( Get f ₁ , t).

In step S106, the attenuation level curve generation unit 30 determines the attenuation level curve GL (f (f ₁ , t)) of the energy attenuation curve GL (f ₁ , t) based on the energy attenuation curve G (f ₁ , t) acquired in step S104. ₁ , t).

In step S108, the smoothing unit 32 performs time averaging processing on the attenuation level curve GL (f ₁ , t) obtained in step S106, and smoothes the attenuation level curve GL (f ₁ , t). Get.

  In step S110, the analysis interval setting unit 34 sets an analysis interval for the smoothed waveform of the attenuation level curve obtained in step S108.

  In step S112, the regression calculation unit 36 calculates a regression line of the smoothed waveform in the analysis section based on the smoothed waveform in the analysis section set in step S110.

In step S114, the decay time calculation unit 38, based on the slope of the regression line calculated in step S112, calculates the decay time of a particular frequency f ₁ of the sound in the target space.

The output unit 40 outputs the decay time of the sound of the specific frequency f ₁ in the target space, calculated by the decay time calculation unit 38 as a result.

  As described above in detail, in the first embodiment, an energy decay curve corresponding to a specific frequency is obtained from the result of the moving Fourier transform of the sound impulse response in the target space, and the regression obtained from the energy decay curve is obtained. The decay time of the sound of a specific frequency in the target space is obtained according to the slope of the straight line. Thereby, the attenuation time of the sound of the specific frequency in object space can be calculated | required, without using the band filter according to the frequency. In addition, by calculating a regression line using a smoothed waveform, the decay time of sound at a specific frequency in the target space can be accurately determined even when the decay level curve obtained from the energy decay curve changes rapidly. Can be sought.

  In the case of using the conventional method, it is necessary to prepare a band filter for each frequency when calculating the decay times of sounds having different frequencies. However, according to the present embodiment, it is possible to obtain the decay time of sound of a desired frequency in the target space by obtaining the moving Fourier transform of the impulse response in the target space.

[Second Embodiment]
Next, a second embodiment of the present invention will be described. Note that the configuration of the decay time analysis system according to the second embodiment is the same as that of the first embodiment, and therefore the same reference numerals are given and description thereof is omitted.

  In the second embodiment, the average of the energy decay curves in a predetermined frequency band is obtained from the result of the moving Fourier transform of the sound impulse response in the target space. And the point which calculates | requires the decay time of the sound of the predetermined | prescribed frequency band in object space differs from 1st Embodiment according to the inclination of the regression line obtained from the energy attenuation | damping curve corresponding to a predetermined | prescribed frequency band.

The frequency setting unit 27 of the second embodiment sets specific frequency bands f _{1 to} f ₂ according to the information received by the receiving unit 10.

For example, the operator of the decay time analyzer 20 sets specific frequency bands f _{1 to} f ₂ for which the decay time is desired, and inputs information regarding the specific frequency bands f _{1 to} f ₂ to the reception unit 10. Then, the frequency setting unit 27 sets specific frequency bands f _{1 to} f ₂ according to the information received by the receiving unit 10.

The energy curve generation unit 28 acquires energy decay curves corresponding to specific frequency bands f _{1 to} f ₂ set by the frequency setting unit 27 from the result of the moving Fourier transform stored in the analysis result storage unit 26.

Specifically, the energy curve generation unit 28 averages both sides of the above formula (2) for specific frequency bands f _{1 to} f ₂ . The energy curve generation unit 28 approximately obtains an energy decay curve for the frequency band corresponding to the integration range by executing the calculation of the following expression (3) (for example, reference documents (Takayuki Hidaka, Yoshinari Yamada, and Takehiko Nakagawa, “A new definition of boundary point between early reflections and latereverberation in room impulse responses.” J. Acoust. Soc. Am. 122 (1), 2007).

  For the energy attenuation curve obtained by the above equation (3), as in the first embodiment, a smoothed waveform of the attenuation level curve of the energy attenuation curve is obtained, an analysis interval is set, and a regression line is obtained. By obtaining, the decay time for a specific frequency band can be obtained.

  In addition, since it is the same as that of 1st Embodiment about the other structure and effect | action of the decay time analysis system which concerns on 2nd Embodiment, description is abbreviate | omitted.

  As described above in detail, in the second embodiment, the average of the energy attenuation curve in the predetermined frequency band is obtained from the result of the moving Fourier transform of the impulse response of the sound in the target space, thereby obtaining the predetermined frequency band. Find the corresponding energy decay curve. Then, the decay time of the sound in a predetermined frequency band in the target space is obtained according to the slope of the regression line obtained from the energy decay curve. Thereby, the attenuation time of the sound in a specific frequency band in the target space can be obtained without using a band filter corresponding to the frequency band.

When using the conventional method, when changing the bandwidth of the frequency band, a band filter corresponding to the frequency band must be prepared. On the other hand, according to the second embodiment, once the moving Fourier transform of the above equation (2) is obtained, it can be dealt with by changing the frequency bands f _{1 to} f ₂ averaged in the above equation (3). A more detailed and multifaceted attenuation analysis becomes possible.

[Third Embodiment]
Next, a third embodiment of the present invention will be described. Note that the configuration of the decay time analysis system according to the third embodiment is the same as that of the first embodiment, and thus the same reference numerals are given and description thereof is omitted.

  In the third embodiment, energy decay curves corresponding to a plurality of different frequency bands are obtained by obtaining an average of the energy decay curves in a plurality of different frequency bands from the result of the moving Fourier transform of the sound impulse response in the target space. Ask. The first or second embodiment is that the sound attenuation time of a plurality of different frequency bands in the target space is obtained according to the slope of the regression line obtained from the energy attenuation curves corresponding to the plurality of different frequency bands. And different.

The frequency setting unit 27 of the third embodiment sets a plurality of different frequency bands f _{11 to} f ₁₂ ,..., F _{n1 to} f _n2 according to the information received by the receiving unit 10.

For example, the operator of the decay time analyzer 20 sets a plurality of different frequency bands f _{11 to} f ₁₂ ,..., F _{n1 to} f _n2 to obtain the decay time, and a plurality of different frequency bands f _{11 to} f ₁₂ ,. .., F _{n1 to} f _n2 are input to the receiving unit 10. Then, the frequency setting unit 27 sets a plurality of different frequency bands f _{11 to} f ₁₂ ,..., F _{n1 to} f _n2 according to the information received by the receiving unit 10.

The energy curve generation unit 28 corresponds to a plurality of different frequency bands f _{11 to} f ₁₂ ,..., F _{n1 to} f _n2 set by the frequency setting unit 27 from the result of the moving Fourier transform stored in the analysis result storage unit 26. Get the energy decay curve.

  Specifically, the energy curve generation unit 28 approximates the energy decay curves Gc (t) for a plurality of different frequency bands by executing the following expressions (4-1) and (4-2). Generate automatically.

  For the energy decay curve Gc (t) obtained by the above equation (4-1), as in the first embodiment, a smoothed waveform of the decay level curve of the energy decay curve is obtained, and the analysis interval is determined. By setting and obtaining a regression line, attenuation times for a plurality of different frequency bands can be obtained.

  As described above in detail, in the third embodiment, a plurality of different frequencies are obtained by obtaining an average of energy decay curves in a plurality of different frequency bands from the result of the moving Fourier transform of the sound impulse response in the target space. An energy decay curve corresponding to the band is obtained. Then, in accordance with the slopes of the regression lines obtained from the energy attenuation curves corresponding to a plurality of different frequency bands, the sound decay times in the plurality of different frequency bands in the target space are obtained. Accordingly, it is possible to obtain the decay time of the sound in which the sounds in the plurality of different frequency bands are superimposed in the target space without using band filters corresponding to the plurality of different frequency bands.

[Fourth Embodiment]
Next, a fourth embodiment of the present invention will be described. Note that the configuration of the decay time analysis system according to the fourth embodiment is the same as that of the first embodiment, and therefore the same reference numerals are given and description thereof is omitted.

  In the fourth embodiment, energy decay curves corresponding to a plurality of different frequencies are obtained by obtaining an average of energy decay curves at a plurality of different frequencies from the result of the moving Fourier transform of the impulse response of the sound in the target space. And the point which calculates | requires the decay time of the sound of several different frequency in object space differs from the 1st-3rd embodiment according to the inclination of the regression line obtained from the energy decay curve corresponding to several different frequency. .

The frequency setting unit 27 of the fourth embodiment sets a plurality of different frequencies f ₁ , f ₂ ,..., F _n according to the information received by the receiving unit 10.

For example, the operator of the decay time analyzer 20, the frequency _{f 1,} f 2 of different plurality it is desired to obtain a decay time, ..., set the _{f n,} the frequency _{f 1,} f 2 a plurality of different, ..., to _{f n} Information is input to the receiving unit 10. Then, the frequency setting unit 27 sets a plurality of different frequencies f ₁ , f ₂ ,..., F _n according to the information received by the receiving unit 10.

The energy curve generation unit 28 is an energy decay curve corresponding to a plurality of different frequencies f ₁ , f ₂ ,..., F _n set by the frequency setting unit 27 from the result of the moving Fourier transform stored in the analysis result storage unit 26. To get.

Specifically, the energy curve generation unit 28 approximates the energy decay curve Gs (t) for a plurality of different frequencies f ₁ , f ₂ ,..., F _n by executing the calculation of the following equation (5). Generate automatically.

FIG. 5 shows an example (logarithmic display) of an energy decay curve Gs (t) for a plurality of different frequencies f ₁ , f ₂ ,..., F _n . 5A is an energy decay curve of a 400 [Hz] sine wave, FIG. 5B is an energy decay curve of a 500 [Hz] sine wave, and FIG. 5C is a 600 [Hz] sine wave. It is an energy decay curve. FIG. 5D is an energy attenuation curve of a sound in which sine waves of 400 [Hz], 500 [Hz], and 600 [Hz] are superimposed. By using the above equation (5), the energy attenuation curve of FIG. 5 (d) is obtained from the energy attenuation curves of FIGS. 5 (a) to 5 (c).

  For the energy decay curve Gs (t) obtained by the above equation (5), as in the first embodiment, a smoothed waveform of the decay level curve of the energy decay curve is obtained, and an analysis interval is set. By calculating the regression line, the decay times for a plurality of different frequencies can be obtained.

  As described above in detail, in the fourth embodiment, the average of the energy decay curves at a plurality of different frequencies is obtained from the result of the moving Fourier transform of the impulse response of the sound in the target space, thereby obtaining a plurality of different frequencies. Find the corresponding energy decay curve. And the decay time of the sound of the several different frequency in object space is calculated | required according to the inclination of the regression line obtained from the energy attenuation | damping curve corresponding to several different frequency. Accordingly, it is possible to obtain the decay time of the sound in which the sounds of the different frequencies are superimposed in the target space without using band filters corresponding to the different frequencies. In addition, the decay time of the sound with overlapping frequencies can be obtained.

<Example of simulation experiment>
Next, the simulation result obtained by the method of this embodiment is shown in FIG.

  The result of FIG. 6 is a simulation result for a reverberation time representing a 60 dB decay time. In the result of FIG. 6, the reverberation time is obtained for each octave band ((a) to (f)). The thin line in FIG. 6 represents the reverberation time of a single frequency component. A thick line represents the reverberation time obtained by the method of the present embodiment. In addition, the reverberation time of the thick line of FIG. 6 is the reverberation time calculated | required from the average energy decay curve of each frequency band. A circle represents the reverberation time obtained by the bandpass filter. As shown in FIG. 6, the reverberation time obtained by the method of the present embodiment corresponds to the reverberation time obtained by the bandpass filter. From this, it can be seen that according to the present embodiment, the attenuation time can be obtained accurately without using a bandpass filter.

  The present invention is not limited to the above-described embodiment, and various modifications and applications can be made without departing from the gist of the present invention.

  For example, in the above-described embodiment, the energy attenuation curve for a specific frequency is obtained from the result of the moving Fourier transform of the impulse response, the attenuation level curve of the energy attenuation curve is calculated, and the attenuation level range of the smoothed waveform of the attenuation level curve Although the case where the slope of the regression line of the smoothed waveform in the analysis section is calculated based on a predetermined analysis section corresponding to the above has been described as an example, the present invention is not limited to this. For example, the decay time of a sound having a specific frequency in the target space may be obtained according to the slope of the regression line of the energy decay curve without obtaining the smoothed waveform of the decay level curve.

  In the above description, the program is stored (installed) in advance in a program storage unit (not shown). However, the program is stored in any recording medium such as a CD-ROM, a DVD-ROM, and a micro SD card. It is also possible to provide it in a recorded form.

DESCRIPTION OF SYMBOLS 10 Reception part 20 Decay time analyzer 22 Response result acquisition part 24 Moving Fourier transform analysis part 26 Analysis result memory | storage part 27 Frequency setting part 28 Energy curve generation part 30 Attenuation level curve generation part 32 Smoothing part 34 Analysis area setting part 36 Regression Calculation unit 38 Decay time calculation unit 40 Output unit 100 Decay time analysis system

Claims (6)

From the result of the moving Fourier transform of the impulse response of the sound in the target space, an energy decay curve corresponding to a specific frequency is obtained,
In accordance with the slope of the regression line obtained from the energy decay curve, find the decay time of the sound of the specific frequency in the target space,
Decay time analysis method. Calculating an attenuation level curve of the energy decay curve;
Based on a predetermined analysis interval corresponding to the level range of attenuation of the smoothed waveform of the attenuation level curve, the slope of the regression line of the smoothed waveform in the analysis interval is calculated,
In accordance with the slope of the calculated regression line of the smoothed waveform, find the decay time of the sound of the specific frequency in the target space,
The decay time analysis method according to claim 1. From the result of the moving Fourier transform, by obtaining an average of the energy attenuation curve in a predetermined frequency band, an energy attenuation curve corresponding to the predetermined frequency band is obtained,
In accordance with the slope of the regression line obtained from the energy decay curve, the decay time of the sound in the predetermined frequency band in the target space is obtained.
The decay time analysis method according to claim 1 or 2. By obtaining an average of the energy attenuation curves in a plurality of different frequency bands or a plurality of different frequencies from the result of the moving Fourier transform, an energy attenuation curve corresponding to the plurality of different frequency bands or the plurality of different frequencies is obtained. ,
In accordance with the slope of the regression line obtained from the energy decay curve, the decay time of the sound of the plurality of different frequency bands or the plurality of different frequencies in the target space is obtained.
The decay time analysis method according to claim 1 or 2. An energy curve generator that obtains an energy decay curve corresponding to a specific frequency from the result of the moving Fourier transform of the impulse response of the sound in the target space;
An attenuation time calculation unit for obtaining an attenuation time of the sound of the specific frequency in the target space according to the slope of the regression line obtained from the energy attenuation curve calculated by the energy curve generation unit;
A decay time analyzer. Computer
An energy curve generator that obtains an energy decay curve corresponding to a specific frequency from the result of the moving Fourier transform of the impulse response of the sound in the target space, and a regression line obtained from the energy decay curve calculated by the energy curve generator A program for functioning as an attenuation time calculation unit for obtaining an attenuation time of the sound of the specific frequency in the target space according to the inclination of the sound.