JP4301099B2 - Impulse response data generation method for sound field simulation - Google Patents

Description

  The present invention relates to a data generation technique for sound field simulation.

  There is a technique for adding sound field information such as reverberation sound when sound is emitted in a specific sound field by simulation to sound data. One of such simulation methods is called a sound ray method.

FIG. 12 is a diagram showing a functional configuration of an apparatus (hereinafter referred to as “sound field simulator 90”) for adding sound field information to sound data by this type of sound ray method. Hereinafter, the operation of the sound field simulator 90 will be described. First, the sound field simulator 90 stores the following data in the storage unit 900 in advance.
(A) Shape data D901: Data indicating the shape of the sound field, which is a collection of coordinate data strings indicating the coordinates of the vertex groups of each boundary surface between the sound field and the outside world.
(B) Sound source position data D902: Data indicating the coordinates of the sound source position in the sound field.
(C) Sound receiving position data D903: Data indicating the coordinates of the sound receiving position in the sound field.
(D) Reflected sound absorption coefficient data D904: When the sound at the reference frequency point (hereinafter referred to as “reference frequency point”) is reflected at the boundary between the sound field and the outside, the intensity and reflection of the sound immediately before reflection. It is data indicating a value obtained by dividing the difference in sound intensity immediately after by the sound intensity immediately before reflection. The reflection sound absorption coefficient data D904 is prepared for each of a plurality of reference frequency points arranged at octave intervals such as 31.3 Hz, 62.5 Hz, 125 Hz,.
(E) Sound data D905: Data indicating the sound before the sound field information is added.

  When the sound is emitted from the sound source position indicated by the sound source position data D902 using the shape data D901, the sound source position data D902, and the sound receiving position data D903, the path calculation unit 901 receives the sound as the sound receiving position data D903. Calculates the route to reach the sound reception position indicated by, and generates route data consisting of coordinate sequences that indicate the sound source position that is the start point of the route, the sound reception position that is the end point, and the reflection position that is the transit point To do.

  The sound path calculated by the path calculation unit 901 is a path obtained by connecting the sound source position and the sound receiving position with a straight line, and the sound emitted from the sound source position is the sound field indicated by the shape data D901 and the outside world. It also includes a path that reaches the sound receiving position after being reflected at the boundary surface. Although there are an infinite number of such routes, the route calculation unit 901 generates route data only for routes that satisfy a predetermined condition, for example, the number of reflections is a certain number or less and the total length of the route is a certain distance or less. A plurality of route data generated by the route calculation unit 901 is stored in the temporary storage unit 906 as a route data group D906.

  The point echo diagram calculation unit 902 receives the sound at the sound receiving position when the sound at the reference frequency point is emitted at the sound source position for each of the reference frequency points using the path data group D906 and the reflection sound absorption coefficient data D904. An echo diagram showing the relationship between the time and intensity of the sound (hereinafter, the echo diagram relating to the sound at the frequency point is called “point echo diagram”) is calculated, and point echo diagram data D907 indicating the echo diagram is generated. FIG. 13 is a diagram illustrating the contents of the point echo diagram data D907. The point echo diagram data D907 generated by the point echo diagram calculation unit 902 is stored in the temporary storage unit 906. Hereinafter, as an example, a method in which the point echo diagram calculation unit 902 generates the point echo diagram data D907 related to the reference frequency point of 125 Hz will be described.

  The point echo diagram calculation unit 902 performs the following processing for each piece of route data included in the route data group D906. First, the point echo diagram calculation unit 902 adds all the distances between the coordinates of adjacent reflection positions included in the route data, and calculates the total length of the route indicated by the route data. Further, the point echo diagram calculation unit 902 counts the number of reflection positions included in the path data. Subsequently, the point echo diagram calculation unit 902 divides the total length of the previously calculated path by the propagation velocity of the sound in the air, so that the sound emitted at the sound source position follows the path indicated by the path data. The time required to reach the sound receiving position (hereinafter referred to as “arrival time”) is calculated. Also, the point echo diagram calculation unit 902 calculates the reciprocal of the value obtained by multiplying the square of the total length of the route by 4π. The numerical value obtained in this way is a numerical value indicating the strength when a sound emitted at a sound source position with a strength “1.0” is received at a sound receiving position after being attenuated by divergence (hereinafter, “ This is referred to as “post-divergence tone intensity”. However, the post-divergence tone intensity does not take into account attenuation due to sound reflection. Since the data indicating the total length of the path calculated as described above, the number of reflection positions, the arrival time, and the intensity value after divergence are not influenced by the frequency, it is also common to reference frequency points other than 125 Hz. Used.

  Subsequently, the point echo diagram calculation unit 902 first counts the numerical value obtained by subtracting from 1 the ratio indicated by the reflection sound absorption coefficient data D904 corresponding to 125 Hz, that is, the attenuation rate, with respect to the post-divergence sound intensity value. Multiply by the number of reflection positions. The numerical value obtained as a result is the sound at the sound receiving position when the sound of 125 Hz emitted at the level of “1.0” at the sound source position reaches the sound receiving position via the reflection point indicated by the route data. Is a numerical value (hereinafter referred to as “attenuated sound intensity value”).

  As described above, the point echo diagram calculation unit 902 can calculate the arrival time of the sound of 125 Hz and the sound intensity value after attenuation for the path indicated by one path data. The point echo diagram calculation unit 902 repeats the same calculation process for all other route data included in the route data group D906. The data group indicating the combination of the arrival time and the attenuated sound intensity value obtained in this way is the point echo diagram data D907 related to the reference frequency point of 125 Hz (see FIG. 13). However, when there are a plurality of post-attenuation sound intensity values corresponding to the same arrival time, the point echo diagram calculation unit 902 uses the sum value of these attenuation sound intensity values as an after-attenuation sound intensity value corresponding to the arrival time. And The above is the processing when the point echo diagram calculation unit 902 generates the point echo diagram data D907.

  The band echo diagram calculation unit 903 convolves the echo waveform indicated by the point echo diagram data D907 related to the reference frequency point of 31.3 Hz with a coefficient sequence of a time waveform having a frequency component shown in FIG. Band echo diagram data D908 indicating an echo diagram relating to a frequency band of 44.2 Hz or less is generated. Similarly, the band echo diagram calculation unit 903 performs, for each of the echo diagrams indicated by the point echo diagram data D907 related to the reference frequency points of 62.5 Hz, 125 Hz,..., FIG. The band echo diagram data D908 relating to the frequency band having the reference frequency point as the center frequency point is generated by convolving the coefficient sequence of the time waveform having the frequency components shown in FIG. The band echo diagram data D908 generated by the band echo diagram calculation unit 903 is stored in the temporary storage unit 906.

  The adding unit 904 adds all the band echo diagram data D908 generated by the band echo diagram calculating unit 903. The data obtained in this way is data that shows a temporal change in the intensity of the sound received at the sound receiving position when sound at any frequency point is pronounced at the sound source position. That is, since the data generated by the adding unit 904 is data that artificially shows the impulse response at the sound receiving position when an impulse is emitted at the sound source position, this is hereinafter referred to as impulse response data D909. The impulse response data D909 generated by the adding unit 904 is stored in the temporary storage unit 906.

  The convolution operation unit 905 convolves the sound data D905 with the coefficient sequence indicated by the impulse response data D909, thereby providing sound field information-added sound data D910 indicating the sound to which the sound field information indicated by the impulse response data D909 is added. Is generated. That is, the sound obtained when the sound data D910 with sound field information is reproduced is reproduced according to the sound data D905 at the sound source position indicated by the sound source position data D902 in the sound field actually indicated by the shape data D901. In this case, the sound is similar to the sound received at the sound receiving position indicated by the sound receiving position data D903.

  This completes the description of the procedure for adding sound field information to sound data by the sound ray method. According to the sound ray method, a similar result can be obtained by calculation processing without actually sounding and recording in a concert hall or the like. At that time, it is easy to obtain results when the sound source position and the sound receiving position are changed variously. Furthermore, according to the sound ray method, it is possible to confirm an acoustic effect in a sound field that does not exist.

For example, Patent Documents 1 and 2 disclose techniques related to the sound ray method.
JP 60-046430 A JP 2000-284788 A

  By the way, the sound obtained by the above-described conventional technology is compared with the sound heard in the actual sound field, and the sound component included in the specific frequency band is cut at the timing when a certain time has elapsed after the sound generation. Accompanies unnaturalness in the sense of hearing. The reason for this unnaturalness is that the echo diagrams relating to adjacent frequency bands are discontinuously joined on the frequency axis. For example, the level of the frequency component of 44.2 Hz of the echo diagram indicated by the band echo diagram data D908 relating to the frequency band below 44.2 Hz and the band echo diagram data D908 relating to the frequency band of 44.2 Hz to 88.4 Hz are indicated. There is a gap that should not actually exist between the level of the frequency components of 44.2 Hz in the echo diagram. This gap causes the above-mentioned unnaturalness in hearing.

  In order to reduce the gap, it is conceivable to increase the resolution on the frequency axis of the echo diagram used to obtain the impulse response. That is, by increasing the number of reference frequency points, the width of the frequency band covered by each band echo diagram data D908 is narrowed. However, when the number of reference frequency points is increased, the amount of data such as the reflection sound absorption coefficient data D904 necessary for generating the point echo diagram data D907 increases, and the effort and cost for preparing these data increases. Further, regarding the data such as the reflection sound absorption coefficient data D904, when a large number of reference data points already exist with respect to a predetermined reference frequency point having an octave interval, if the number of reference frequency points is increased, those libraries are stored. There is an inconvenience that data cannot be used.

  In view of the above situation, an object of the present invention is to provide means for obtaining impulse response data that provides a more natural audibility without increasing the number of reference frequency points.

Hereinafter, a method for increasing the resolution on the frequency axis of an echo diagram used for obtaining an impulse response according to the present invention will be described. First, when the Fourier transform of the impulse waveform Im (t) is a (jω), the impulse response waveform Im (t) can be expressed as the following (Equation 1).

If the frequency transfer characteristic from the sound source position to the sound receiving position in the sound field is F (jω), the impulse response waveform Res (t) observed at the sound receiving position when the impulse waveform Im (t) is generated at the sound source position. Can be expressed as the following (Formula 2).

Here, if the impulse response waveform obtained in the actual sound field is Fourier transformed, F (jω) can be obtained. However, actually measuring the impulse response waveform in an actual sound field requires cost and labor, and there is also a problem that it is impossible to measure in a sound field that does not exist. Therefore, it is necessary to approximate the impulse response waveform by simulation using a computer. For this purpose, first, the impulse waveform Im (t) is approximated as in the following (formula 3). Here, in the following (Expression 3), the limit where n is set to ∞ and Δω is close to 0 is (Expression 2).

The response waveform Res _k (t) corresponding to one spectrum a (jkΔω) e ^jkΔωt constituting the impulse waveform Im (t) is as shown in the following (formula 4).

The impulse response waveform Res (t) can be considered as the sum of the response waveforms Res _k (t) corresponding to all the spectra a (jkΔω) e ^jkΔωt constituting the impulse waveform Im (t). 5).

In the above (Equation 3) and (Equation 4), F (jkΔω) assumes various sound propagation paths from the sound source position to the sound reception position, and calculates the delay time and attenuation rate in each propagation path, For example, it can be obtained as follows.

・・・（式６） Therefore, the sum of the transfer functions corresponding to all the propagation paths from the sound source position to the sound receiving position, that is, the transfer function from the sound source position to the sound receiving position can be expressed as the following (Equation 6). The echo diagram described above corresponds to this equation (6).

... (Formula 6)

Here, in order to reproduce more natural sound field information, it is necessary to increase the frequency resolution of the impulse waveform Im (t), that is, to increase the number n of a (jkΔω) e ^jkΔωt . On the other hand, increasing the frequency resolution of F (jω) in accordance with the frequency resolution of the impulse waveform Im (t) and obtaining the above (formula 6) for the number of angular frequencies ω is that the attenuation rate σ and the delay time τ Since it is necessary to prepare parameters for calculating each angular frequency ω, it is not easy and requires a huge amount of calculation. Therefore, a means for reducing the frequency resolution of F (jω) with respect to the frequency resolution of the impulse waveform Im (t) is required.

この結果、従来技術は、上述した不都合を生んだ。 In the prior art, the frequency resolution of the impulse waveform Im (t) is used by using F (jkΔω) related to the reference angular frequency ω as it is as a transfer function corresponding to a (jkΔω) e ^jkΔωt related to the neighboring angular frequency ω. On the other hand, the frequency resolution of F (jω) was lowered. For example, it is as follows. However, “→” indicates that the right is a transfer function associated with the left.

As a result, the prior art has caused the above-mentioned disadvantages.

In contrast, in the present invention, the transfer function of a (jkΔω) e ^jkΔωt related to each angular frequency ω is expressed as F (jkΔω) related to the reference angular frequency ω located on the low frequency side and the high frequency side, respectively. It was determined by interpolation. For example, it is as follows.

More specifically, in the present invention, a plurality of feature data indicating the characteristics of the sound field, sound source position data indicating the sound source position in the sound field, and sound receiving position data indicating the sound receiving position in the sound field are used. When each sound of a reference frequency is generated at the sound source position, an echo diagram representing a temporal change in the sound intensity level of the sound reaching the sound receiving position is calculated for each of the plurality of reference frequencies, The number of interpolation frequencies between a plurality of reference frequencies is a predetermined fixed number, a specified number, a number associated with a selected music genre, directivity of a sound source in the sound field, and the reference frequency Calculate the frequency component included in the time waveform of the sound before the reverberation sound is added, the number set based on the air sound absorption coefficient and the reflection sound absorption coefficient related to the reference frequency, and between the adjacent reference frequencies Number set in accordance with the magnitude of the amplitude values of the frequency components contained in the frequency band, to determine the one of the plurality of reference frequencies, predetermined at the time of sound the sound of the reference frequency at the sound source position The directivity data in which the sound intensity level at the determined position is associated is acquired, and the reference frequency in the acquired directivity data is set to x in the coordinates where the frequency is the x coordinate of the logarithmic axis and the sound intensity level is the y coordinate. The number of interpolation frequencies determined by calculating a line passing through the coordinates corresponding to the sound intensity level of each reference frequency at the same time with the sound intensity level in the coordinates and the directivity data as the y coordinate calculate the echo diagram in the dividing the line into equal parts, each of the one or more interpolation frequency corresponding to the coordinates of the dividing position between the coordinates corresponding to a plurality of reference frequency based on And, for each of the plurality of reference frequencies and the interpolation frequency, a coefficient sequence representing a time waveform having a positive frequency component only in a passband including the reference frequency or the interpolation frequency as a center frequency and including the center frequency, A convolution is performed on the echo diagram corresponding to the reference frequency and the interpolation frequency, and the echo diagram subjected to the convolution related to each of the plurality of reference frequencies and the interpolation frequency is added to represent the sound transfer characteristics in the sound field. Generate impulse response waveform data.

  In a preferred embodiment, the upper limit of one pass band including one center frequency and another pass including another center frequency adjacent to the one center frequency and located higher than the one center frequency. The lower limit frequency of the band matches.

  As a preferred mode for determining the coefficients for the above interpolation, that is, the values of b and c, parameters for calculating the attenuation rate σ and the delay time τ with respect to the reference angular frequency ω are interpolated, and the interpolated parameters are used. It is conceivable that the above b and c are determined by calculating the attenuation rate σ and the delay time τ with respect to the interpolated angular frequency ω.

  More specifically, using the feature data including information indicating the sound absorption coefficient of the boundary surface of the sound field at each of the plurality of reference frequencies, the impulse response data generation method uses each of the plurality of reference frequencies. By interpolating the sound absorption rate at each of the plurality of reference frequencies to calculate a sound absorption rate at each of the one or more interpolation frequencies, the feature data, the sound source position data, the sound reception position data, and the one or more An echo diagram representing a temporal change in the level of the sound that reaches the sound receiving position when sound of each of the one or more interpolation frequencies is generated at the sound source position using the sound absorption rate at each of the interpolation frequencies. For each of the one or more interpolation frequencies, so that an echo die at each of the one or more interpolation frequencies between the plurality of reference frequencies is calculated. To calculate the grams.

  By the way, as another means for solving the problem that the echo diagrams related to adjacent frequency bands are discontinuously joined on the frequency axis, a coefficient sequence of a time waveform having a frequency component that cross-fades on the frequency axis, It is conceivable that the transfer function related to the reference angular frequency ω is convolved with an echo diagram developed on the time axis.

  More specifically, each of a plurality of reference frequencies using feature data indicating the characteristics of the sound field, sound source position data indicating the sound source position in the sound field, and sound receiving position data indicating the sound receiving position in the sound field. An echo diagram representing a temporal change in the level of the sound reaching the sound receiving position when the sound is generated at the sound source position is calculated for each of the plurality of reference frequencies, and each of the plurality of reference frequencies is calculated. An echo diagram corresponding to the reference frequency is a coefficient sequence representing a time waveform having a frequency component that attenuates by drawing a slope toward the low frequency side and the high frequency side of each center frequency with the reference frequency as the center frequency. And transmitting the convolutional echo diagram for each of the plurality of reference frequencies to add a sound transmission characteristic in the sound field. Generating an impulse response waveform data representing the.

  Moreover, in a preferable aspect, the band in which the frequency component of the time waveform used in the convolution takes a positive value has an overlapping portion between adjacent bands. In a more preferred aspect, the sum of frequency components in the overlapping portion is 1.

  Furthermore, by combining the calculation of the echo diagram related to the interpolation frequency described above and the convolution of the coefficient sequence of the time waveform having the frequency component that cross-fades on the frequency axis, the echo diagram related to the adjacent frequency band can be made smoother on the frequency axis. It becomes possible to join to.

More specifically, each of a plurality of reference frequencies using feature data indicating the characteristics of the sound field, sound source position data indicating the sound source position in the sound field, and sound receiving position data indicating the sound receiving position in the sound field. Calculating an echo diagram for each of the plurality of reference frequencies representing a temporal change in the sound intensity level of the sound that reaches the receiving position when the sound is generated at the sound source position; The number of interpolation frequencies between the reference frequencies is a predetermined fixed number, a specified number, the number associated with the selected music genre, the directivity of the sound source in the sound field, and the air sound absorption related to the reference frequency The frequency component included in the time waveform of the sound before the reverberation sound is added, the number set based on the reflection sound absorption coefficient relating to the reference frequency and the reference frequency, and the frequency between the adjacent reference frequencies is calculated. Number set in accordance with the magnitude of the amplitude values of the frequency components contained in the band, decides in any one of the plurality of reference frequency, predefined when pronouncing the sound of the reference frequency at the sound source position The directivity data in which the sound intensity level at the selected position is associated with each other, and the reference frequency in the acquired directivity data is the x coordinate in the coordinates where the frequency is the x coordinate of the logarithmic axis and the sound intensity level is the y coordinate. The sound intensity level in the directivity data is defined as a y-coordinate, and a line passing through the plurality of reference frequencies and a coordinate corresponding to the sound intensity level of each reference frequency at the same time is calculated, and the number of interpolation frequencies determined is calculated. based by dividing the line between the coordinates corresponding to a plurality of reference frequencies equally, calculated echo diagram in each of one or more interpolation frequency corresponding to the coordinates of the division position For each of the plurality of reference frequencies and the one or more interpolation frequencies, a frequency component that attenuates by drawing a slope toward the low frequency side and the high frequency side of each center frequency with the reference frequency or the interpolation frequency as the center frequency. A coefficient sequence representing a time waveform having a frequency is convolved with an echo diagram corresponding to the reference frequency and the interpolation frequency, and the convolved echo diagram for each of the plurality of reference frequencies and the one or more interpolation frequencies is added. As a result, impulse response waveform data representing the sound transfer characteristics in the sound field is generated.

  According to the impulse response data generation method having such a configuration, impulse response data with reduced discontinuities on the frequency axis is generated without increasing the resolution of the reference frequency.

  The present invention also provides an apparatus for realizing the impulse response data generation method described above by data processing. Furthermore, the present invention provides a program showing a procedure for realizing each of the above impulse response data generation methods by causing a computer to perform predetermined data processing.

  According to the method of the present invention, it is possible to obtain impulse response data that provides a more natural audibility without increasing the number of reference frequency points.

[1. First Embodiment]
The data processing apparatus according to the first embodiment of the present invention (hereinafter referred to as “sound field simulator 10”) generates impulse response data for adding sound field information to sound data, and generates the generated impulse response data. Is a device that generates sound data indicating a sound to which sound field information is added by performing a process of convolving the sound data with the sound data.

[1.1. Configuration of sound field simulator]
FIG. 1 is a diagram illustrating a hardware configuration of the sound field simulator 10. The sound field simulator 10 is a CPU (Central Processing Unit) 101 that performs various data processing and controls other components as in a general computer, and a BIOS (Basic Input Output System) that performs basic processing on the CPU 101. ) And the like, a RAM (Random Access Memory) 103 for temporarily storing data in the data processing of the CPU 101, and an application program for realizing the functions of the sound field simulator 10 (hereinafter, “ A sound field simulation program), an HD (Hard Disk) 104 for storing various data, a screen for prompting the user to select sound data in accordance with instructions from the CPU 101, etc. A display system 105 for displaying, an operation unit 106 such as a keyboard for transmitting a signal according to a user's operation, a D / A (Digital to Analog) converter, a sound system 107 and an external device that perform sound generation based on sound data. Input / output I / F (Interface) 108 for transmitting / receiving data to / from the.

  The CPU 101 realizes various functions of the sound field simulator 10 by executing a sound field simulation program. FIG. 2 is a block diagram showing a functional configuration of the sound field simulator 10. However, a display unit and an operation unit for the sound field simulator 10 to exchange information with the user, a communication unit for performing data communication with an external device, and the like are the same as those in a general information processing device. 2 is omitted in FIG. Each component of the sound field simulator 10 uses various data included in the storage unit 900 and the temporary storage unit 906. However, since all of them are complicated, the use of main data in FIG. Only the relationships are indicated by arrows. The function of the sound field simulator 10 has a common part with the function of the sound field simulator 90 shown in FIG. Therefore, in FIG. 2, the same reference numerals are given to the functional components that are the same as or similar to those in FIG. 12.

The storage unit 900 of the sound field simulator 10 stores the following data in advance as the same or similar to the various data stored in the storage unit 900 of the sound field simulator 90. However, in the following description, the unit of the numerical value indicated by the coordinate data is assumed to be meters. The numerical values of the data shown below are for explanation, and do not necessarily mean actual numerical values.
(A) Shape data D901: data indicating the shape of the sound field and the material of each boundary surface between the sound field and the outside world, a coordinate data string indicating the coordinates of the vertex group of the boundary surface and the material data indicating the material of the boundary surface Is a collection of data (hereinafter referred to as “boundary surface data”). FIG. 3 shows an example of the shape data D901.
(B) Sound source position data D902: Data indicating the sound source position included in the sound field and the direction in which the sound source emits sound, and is a combination of the coordinate data of the sound source position and vector data indicating the direction in which the sound is emitted. For example, the sound source position data D902 has a data format of “(6, 5, 3), (0, 0.87, 0.5)”, and in this case, the sound source indicated by the coordinate data (6, 5, 3). It shows that a sound is emitted from the position in the direction indicated by the vector data (0, 0.87, 0.5).
(C) Sound reception position data D903: Coordinate data indicating the sound reception position included in the sound field.
(D) Reflected sound absorption coefficient data D904: When the sound at each reference frequency point is reflected at the boundary between the sound field and the outside, the difference between the sound intensity immediately before reflection and the sound intensity immediately after reflection is calculated as The sound absorption coefficient data indicating the value divided by the sound intensity of the sound is associated with the material data indicating the material of the boundary surface and the incident angle data indicating the incident angle of the sound (hereinafter referred to as “sound absorption coefficient data by material incident angle”). "). FIG. 4 shows an example of the reflection sound absorption coefficient data D904.
(E) Sound data D905: Data indicating the sound before the sound field information is added.

The storage unit 900 of the sound field simulator 10 further stores the following data in advance.
(F) Sound source name data D200: Data indicating the name of the sound source.
(G) Temperature and humidity data D201: Data indicating the temperature and humidity of the air in the sound field.
(H) Interpolation number template data group D202: Interpolation number data indicating the number of frequency points (hereinafter referred to as “interpolation frequency points”) for interpolating between adjacent reference frequency points in order from the low frequency side, such as music A group of data (hereinafter referred to as “interpolation number template data”) associated with genre data indicating a classification name. FIG. 5 shows an example of the interpolation number template data group D202. For example, the interpolation number template data shown in the first row of FIG. 5 indicates the number of interpolation frequency points suitable for adding sound field information to sound data indicating music belonging to the genre of “Rock”. (1, 1,..., 3) indicated by the numerical data are frequency bands “31.3 Hz to 62.5 Hz”, “62.5 Hz to 125 Hz”,..., “8 kHz to 16 kHz” in order. It corresponds to. Interpolation number data included in each interpolation number template data is created so as to indicate a larger number of interpolations in a frequency band that significantly includes those features, depending on the genre of the music.
(I) Directivity data D203: Vector data indicating the position of each point on the sphere centered on the reference point and having a unit distance (for example, 1 meter) as a radius, and the sound at the reference frequency point by a specific sound source at the reference point. Data (hereinafter referred to as “directivity”) associated with sound intensity data indicating the intensity of sound received at a position on the spherical surface when sounding in a reference direction, for example, (0, 1, 0). A collection of element data). The directivity data D203 indicates the directivity of the sound source regarding the sound at each reference frequency point as a whole. The directivity data D203 is prepared for each of a plurality of sound sources. FIG. 6 shows an example of the directivity data D203. The directivity element data in the first row in FIG. 6 is sounded by the “speaker SP1” at a position 1 meter from the sound source position in the direction indicated by the vector data (0, 1, 0), that is, the reference direction. .3 Hz sound is received with an intensity of “0.97”. Further, the directivity element data in the second row in FIG. 6 is pronounced by the “speaker SP1” at a position 1 meter from the sound source position in a direction slightly inclined in the positive x-axis direction from the reference direction. This indicates that a 3 Hz sound is received with an intensity of “0.95”. The type of sound source is not limited to a speaker, and may be a musical instrument.
(J) Air sound absorption coefficient data D204: When the sound at each reference frequency point propagates by a unit distance (for example, 1 meter), the difference between the sound intensity before propagation and the sound intensity after propagation is calculated as follows. A collection of data (hereinafter referred to as "sound absorption coefficient data by temperature and humidity") in which sound absorption coefficient data indicating a value divided by sound intensity is associated with temperature data indicating air temperature and humidity data indicating humidity. is there. FIG. 7 shows an example of the air sound absorption rate data D204.

  Among the data (a) to (j), (a) shape data D901, (b) sound source position data D902, (c) sound receiving position data D903, (e) sound data D905, and (f) sound source name. The data D200 and (g) temperature / humidity data D201 can be freely changed or selected according to the sound field environment or the like that the user wants to perform simulation. As for other data, the user can use the data already created as a library in addition to the data created by the user himself / herself.

[1.2. Sound field simulator functions and operations]
Hereinafter, functions and operations of the sound field simulator 10 will be described. First, the route calculation unit 901 generates a route data group D906 using the shape data D901, the sound source position data D902, and the sound receiving position data D903. The method by which the route calculation unit 901 generates the route data group D906 is the same as in the sound field simulator 90.

When the route data group D906 is stored in the temporary storage unit 906, the sound field simulator 10 subsequently performs data interpolation processing. Therefore, the sound field simulator 10 displays a screen that prompts the user to select one of the following options regarding the method for determining the number of interpolations.
(A) Fixed number “3”
(B) Individual specification (c) Template selection (d) Automatic setting based on reference data (e) Automatic setting based on sound data

  The interpolation number determination unit 201 determines the number of interpolation frequency points to be interpolated between reference frequency points in accordance with a user operation on the screen showing the above options. When the option (a) is selected by the user, the interpolation number determination unit 201 creates interpolation number data D205 having the contents (3, 3,..., 3). The numerical values included in the interpolation number data D205 are “31.3 Hz to 62.5 Hz” and “62.5 Hz to 125 Hz, respectively, similarly to the interpolation number data included in the interpolation number template data group D202 (see FIG. 5) described above. ,..., “8 kHz to 16 kHz” indicates the number of interpolation frequency points for interpolating each frequency band.

  When the option (b) is selected by the user, the sound field simulator 10 further displays a screen that prompts the user to input the number of interpolation frequency points in each frequency band. When the user inputs a numerical value in accordance with the screen, the interpolation number determination unit 201 creates data indicating a string of numerical values input by the user as interpolation number data D205. Here, the user can input “0” as the number of interpolation frequency points for all adjacent reference frequency points. In that case, a series of interpolation processing described below is skipped.

  When the above option (c) is selected by the user, the sound field simulator 10 further displays a screen that prompts the user to select one from a list indicating the contents of the genre data included in the interpolation number template data group D202. indicate. When the user selects one of the genres according to the screen, the interpolation number determination unit 201 copies a copy of the interpolation number data of the interpolation number template data including the genre data indicating the genre selected by the user to the interpolation number data D205. Create as.

  When the option (d) is selected by the user, the interpolation number determination unit 201 calculates and calculates an index indicating the degree of difference for each of the directivity data D203, the air sound absorption coefficient data D204, and the reflection sound absorption coefficient data D904. The number of interpolation frequency points between the respective reference frequency points is determined as follows according to the index.

  First, the interpolation number determination unit 201 indicates an index (hereinafter referred to as a degree of difference) between the directivity data D203 (see FIG. 6) corresponding to the sound source name indicated by the sound source name data D200 and related to adjacent reference frequency points. , Called “directivity difference index”). For example, when the sound source name data D200 indicates “speaker SP1”, the interpolation number determination unit 201 uses the same vector for the 31.3 Hz directivity data D203 and 62.5 Hz directivity data D203 corresponding to “speaker SP1”. By adding the square of the difference between the values indicated by the sound intensity data included in the directional element data including the data with respect to all the directional element data, the directivity difference index between the directional data D203 is calculated. The interpolation number determination unit 201 similarly calculates a directivity difference index for other combinations of adjacent reference frequency points.

  Subsequently, the interpolation number determination unit 201 includes the temperature data and the humidity data indicating the temperature and humidity indicated by the temperature / humidity data D201 among the air sound absorption coefficient data D204 (see FIG. 7) regarding each reference frequency point. The sound absorption coefficient data of the sound absorption coefficient data is read. The interpolation number determination unit 201 indicates an index indicating the degree of difference between the sound absorption coefficients indicated by the sound absorption coefficient data thus read out and those related to adjacent reference frequency points (hereinafter referred to as “air sound absorption coefficient difference index”). ) Is calculated. For example, when the temperature / humidity data D201 indicates “10 ° C.” and “10%”, the interpolation number determination unit 201 determines that the temperature data is “10 ° C.” from each of the air sound absorption coefficient data D 204 of 31.3 Hz and 62.5 Hz. And the sound absorption coefficient data of the temperature-humidity absorption coefficient data whose humidity data is “10%” is read out, and the square of the difference between the values indicated by the read sound absorption coefficient data is calculated as an air absorption coefficient difference index. The interpolation number determination unit 201 calculates the air absorption coefficient difference index similarly for other combinations of adjacent reference frequency points.

  Subsequently, the interpolation number determination unit 201 extracts the sound absorption coefficient data for each material incident angle including the material data included in the shape data D901 from each of the reflection sound absorption coefficient data D904 (see FIG. 4) regarding the adjacent reference frequency points. Then, an index indicating the degree of difference in the sound absorption coefficient data included in the extracted sound absorption coefficient data for each incident angle of the material (hereinafter referred to as “reflected sound absorption coefficient difference index”) is calculated. For example, when the shape data D901 includes material data indicating “concrete”, the interpolation number determination unit 201 determines that the material data is “concrete” from each of the reflected sound absorption coefficient data D904 of 31.3 Hz and 62.5 Hz. Extracts sound absorption coefficient data for a certain material incident angle. The interpolation number determination unit 201 is included in the sound absorption coefficient data by material incident angle extracted from the reflection sound absorption coefficient data D904 at 31.3 Hz and the sound absorption coefficient data by material incident angle extracted from the reflection sound absorption coefficient data D904 at 62.5 Hz. The square of the difference in sound absorption coefficient indicated by the sound absorption coefficient data is calculated between the data having the same incident angle data, and all of them are added. The value thus calculated is an index indicating the degree of difference in the reflection sound absorption coefficient regarding “concrete”.

  The interpolation number determination unit 201 similarly calculates an index for other material data. Further, the interpolation number determination unit 201 calculates the area of the boundary surface indicated by the coordinate data string for each of the materials indicated by the material data included in the shape data D901. The interpolation number determination unit 201 calculates a weighted average value of an index indicating the degree of difference in the reflection sound absorption coefficient regarding each material data, using the area of the boundary surface regarding each material data as a weight. The weighted average value thus calculated is a reflection sound absorption coefficient difference index related to the reflection sound absorption coefficient data D904 of 31.3 Hz and 62.5 Hz. The interpolation number determination unit 201 similarly calculates the reflection sound absorption coefficient difference index for other combinations of adjacent reference frequency points.

  Subsequently, the interpolation number determination unit 201 calculates the average value of the directivity difference index, the air sound absorption rate difference index, and the reflection sound absorption rate difference index calculated as described above for each of the adjacent reference frequency points. At this time, the interpolation number determining unit 201 uses the directivity data D203, the air sound absorption rate data D204, and the reflection sound absorption rate data D904 to calculate the shape of the echo diagram calculated by the point echo diagram calculation unit 902 using them. The weighted average value of the directivity difference index, the air sound absorption coefficient difference index, and the reflection sound absorption coefficient difference index may be calculated using a weight corresponding to the magnitude of the influence. When the index calculated in this way is large, there is a large difference in the directivity data D203 and the like related to adjacent reference frequency points. Therefore, the interpolation number determination unit 201 sets a large numerical value as the number of interpolation frequency points between the reference frequency points. decide. On the other hand, when the calculated index is small, the interpolation number determination unit 201 determines a small numerical value as the interpolation frequency point between the reference frequency points. The interpolation number determination unit 201 creates data indicating the numerical sequence determined as described above as interpolation number data D205.

  When the option (e) is selected by the user, the interpolation number determination unit 201 performs Fourier transform on all or part of the time waveform of the sound indicated by the sound data D905, and calculates a frequency component included in the time waveform. . The interpolation number determination unit 201 determines a large numerical value as the number of interpolation frequency points according to the magnitude of the amplitude value of the frequency component included in the frequency band between adjacent reference frequency points. The interpolation number determination unit 201 creates data indicating the determined numerical sequence as interpolation number data D205.

  The interpolation number data D205 created by the interpolation number determination unit 201 as described above is stored in the temporary storage unit 906. When the interpolation number data D205 is stored in the temporary storage unit 906, the interpolation directivity calculating unit 202, the interpolated air sound absorption rate calculating unit 203, and the interpolated reflection sound absorption rate calculating unit 204 are respectively directed to the directivity data D203 and the air sound absorption rate data D204. Then, the numerical values at the reference frequency point indicated by the reflection sound absorption coefficient data D904 are interpolated by the number indicated by the interpolation number data D205 to generate the interpolated directivity data D206, the interpolated air sound absorption coefficient data D207, and the interpolated reflection sound absorption coefficient data D208. . Hereinafter, those processes will be described.

  Interpolation directivity calculation unit 202 reads directional element data having the same vector data from directivity data D203 (see FIG. 6) regarding each reference frequency point corresponding to the sound source name indicated by sound source name data D200. A spline curve passing through a point having a numerical value indicated by the sound intensity data included in the directivity element data as the y coordinate and a corresponding reference frequency point as the x coordinate is calculated. Subsequently, the interpolation directivity calculation unit 202 is a point on the calculated spline curve, and when the x axis is logarithmically displayed, each of the x coordinates indicating each reference frequency point is represented by the number indicated by the interpolation number data D205. Calculate the coordinates of the points to be internally divided.

  For example, when the initial number indicated by the interpolation number data D205 is “3”, the interpolation directivity calculation unit 202 divides 31.3 Hz and 62.5 Hz into four equal parts in the logarithmic display, that is, 37.2 Hz. , 44.2 Hz and 52.6 Hz are obtained as the coordinates of the points on the spline curve. The x-coordinates of these points (hereinafter referred to as “interpolation points”) are interpolation frequency points, and the y-coordinates indicate estimated sound intensity values for the interpolation frequency points. When the interpolation directivity calculation unit 202 calculates the coordinates of the interpolation points as described above with respect to all directivity element data included in the directivity data D203, the vector data and the y coordinates are calculated for each x coordinate, that is, for each interpolation frequency point. That is, a collection of data in which sound intensity data indicating an estimated value of sound intensity is associated is created. The data thus created is the interpolation directivity data D206. The structure of the interpolated directivity data D206 is the same as the directivity data D203 (see FIG. 6) corresponding to one sound source name. The interpolation directivity data D206 created by the interpolation directivity calculation unit 202 is stored in the temporary storage unit 906.

  Here, the use of a spline curve is merely an example, and the interpolation directivity calculation unit 202 calculates another type of curve or a point on a straight line connecting points corresponding to adjacent reference frequency points as an interpolation point. May be. The same applies to the creation of the interpolated air sound absorption coefficient data D207 and the interpolated reflection sound absorption coefficient data D208 below.

  The interpolated air sound absorption rate calculation unit 203 obtains the sound absorption rate by temperature / humidity including temperature data and humidity data indicating the temperature and humidity indicated by the temperature / humidity data D201 from the air sound absorption rate data D204 (see FIG. 7) for each reference frequency point. The data is read out, and a spline curve is calculated that passes through a point having the numerical value indicated by the sound absorption coefficient data included in the read sound absorption coefficient data by temperature and humidity as the y coordinate and the corresponding reference frequency point as the x coordinate. Subsequently, the interpolation air sound absorption rate calculation unit 203 calculates the coordinates of the number of interpolation points indicated by the interpolation number data D205 as in the case of the interpolation directivity calculation unit 202 described above. The y-coordinate of the coordinates thus calculated indicates the estimated value of the air sound absorption rate for the interpolation frequency point.

  The interpolation air sound absorption rate calculation unit 203 creates data indicating the y coordinate of the interpolation point as interpolation air sound absorption rate data D207 for the interpolation frequency point indicated by the x coordinate of the interpolation point calculated as described above. Here, the interpolated air sound absorption coefficient data D207 relating to each interpolation frequency point includes only one sound absorption coefficient data indicating an estimated value of the sound absorption coefficient at the temperature and humidity indicated by the temperature / humidity data D201. The interpolated air sound absorption rate data D207 created by the interpolated air sound absorption rate calculating unit 203 is stored in the temporary storage unit 906.

  Interpolated reflection sound absorption coefficient data D208 extracts sound absorption coefficient data by material incident angle including material data included in shape data D901 from reflection sound absorption coefficient data D904 (see FIG. 4) for each reference frequency point. The interpolated reflection sound absorption coefficient calculation unit 204 reads out, for each reference frequency point, the same material data and incident angle data from the extracted sound absorption coefficient data by material incident angle, and is included in the read sound absorption coefficient data by material incident angle. A spline curve passing through a point having the numerical value indicated by the sound absorption coefficient data as the y coordinate and the corresponding reference frequency point as the x coordinate is calculated. Subsequently, the interpolation reflection sound absorption coefficient calculation unit 204 calculates the coordinates of the number of interpolation points indicated by the interpolation number data D205, as in the case of the interpolation directivity calculation unit 202 described above. The y coordinate of the interpolation point thus calculated is an estimated value of the reflection sound absorption coefficient when the sound at the interpolation frequency point is reflected at the incident angle indicated by the incident angle data with respect to the boundary surface of the material indicated by the material data. Indicates.

  The interpolated reflection sound absorption rate calculating unit 204 calculates the coordinates of the interpolation points as described above for all the material incident angle sound absorption rate data extracted previously, and calculates the material data and the x-coordinates, that is, for each interpolation frequency point. A collection of data in which the incident angle data is associated with the y coordinate, that is, the sound absorption coefficient data indicating the estimated value of the reflection sound absorption coefficient is created. The data thus created is the interpolated reflection sound absorption coefficient data D208. The structure of the interpolated reflection sound absorption coefficient data D208 is the same as that of the reflection sound absorption coefficient data D904 (see FIG. 4). However, the interpolated reflection sound absorption coefficient data D208 includes only the sound absorption coefficient data for each material incident angle related to the material data included in the shape data D901. The interpolated reflection sound absorption coefficient data D208 created by the interpolated reflection sound absorption coefficient calculating unit 204 is stored in the temporary storage unit 906. The above is a series of interpolation processes in the sound field simulator 10.

  As described above, when the interpolated directivity data D206, the interpolated air sound absorption coefficient data D207, and the interpolated reflection sound absorption coefficient data D208 are stored in the temporary storage unit 906, the point echo diagram calculation unit 902 includes the path data group D906, the sound source name, and so on. Point echo diagram data D907 for each reference frequency point is created using data D200, temperature / humidity data D201, directivity data D203 for each reference frequency point, air sound absorption coefficient data D204, and reflection sound absorption coefficient data D904. Hereinafter, a procedure in which the point echo diagram calculation unit 902 creates the point echo diagram data D907 will be described.

  First, the point echo diagram calculation unit 902 determines the sound intensity corresponding to the direction emitted from the sound source position for each of the sound paths indicated by the path data group D906 based on the directivity data D203 (see FIG. 6). . Therefore, the point echo diagram calculation unit 902 calculates a difference vector between a vector indicating a reference direction in the directivity data D203, that is, (0, 1, 0) and a vector indicated by vector data included in the sound source position data D902. Ask for. Subsequently, the point echo diagram calculation unit 902 calculates, for each piece of route data included in the route data group D906, a unit vector of a direction vector having a first coordinate included in the route data as a start point and a second coordinate as an end point. . Furthermore, the point echo diagram calculation unit 902 adds the previously obtained difference vector to the unit vector thus calculated. The vector obtained as a result is that the sound that follows the path indicated by the path data is generated from the sound source position when the direction in which the sound indicated by the sound source position data D902 is made to coincide with the reference direction in the directivity data D203. Indicates the direction in which it is emitted. Therefore, the point echo diagram calculation unit 902 searches the directivity data D203 corresponding to the sound source name indicated by the sound source name data D200 for directivity element data including vector data indicating the vector thus calculated. The numerical value indicated by the sound intensity data included in the directivity element data searched in this way is used as the initial value of the sound intensity in the route. Hereinafter, the initial value of the tone strength is referred to as “sound strength value 1”.

  Subsequently, the point echo diagram calculation unit 902 reads temperature-humidity-specific sound absorption coefficient data including temperature data and humidity data indicating the temperature and humidity indicated by the temperature / humidity data D201 from the air sound absorption coefficient data D204 (see FIG. 7). The numerical value indicated by the sound absorption coefficient data included in the read sound absorption coefficient data by temperature and humidity is multiplied by the numerical value indicating the total length of the path indicated by the path data. The numerical value obtained by the multiplication indicates the ratio of the sound intensity attenuated by the air when the sound propagates through the total extension. Therefore, the point echo diagram calculation unit 902 multiplies the previously obtained tone intensity value 1 by a numerical value obtained by subtracting the numerical value obtained by multiplication from 1. The numerical value obtained by the multiplication indicates the sound intensity at the sound receiving position when only the directivity and the air sound absorption factor are considered. Hereinafter, the value indicating the tone strength is referred to as “sound strength value 2”.

  Subsequently, for each of the reflection positions indicated by the path data, the point echo diagram calculation unit 902 calculates the incident angle at the reflection position with respect to the boundary surface indicated by the shape data D901 and specifies the material of the boundary surface. For example, when the route data is “... (6, 9, 0), (6, 0, 9),...” And the shape data D901 includes the boundary surface data shown in FIG. , 0, 9) indicates the reflection position on the boundary surface indicated by the coordinate data string of the boundary surface data in the first row in FIG. 3, and the incident angle is calculated to be 45 degrees. The material corresponding to the reflection position is specified as “concrete”.

  Subsequently, the point echo diagram calculation unit 902 searches the reflection sound absorption coefficient data D904 (see FIG. 4) for the material incident angle sound absorption coefficient data including the material data indicating the specified material and the incident angle data indicating the calculated incident angle. Then, the reflection sound absorption coefficient indicated by the sound absorption coefficient data included in the searched sound absorption coefficient data by material incident angle is specified. The point echo diagram calculation unit 902 subtracts the reflection sound absorption coefficient specified as described above from 1. The point echo diagram calculation unit 902 similarly specifies the reflection sound absorption coefficient for each reflection position, and subtracts each from 1. The point echo diagram calculation unit 902 multiplies all of the numerical values thus calculated by the tone intensity value 2 calculated previously. The numerical value obtained by the multiplication indicates the sound intensity at the sound receiving position when only the directivity, the air sound absorption coefficient, and the reflection sound absorption coefficient are considered. Hereinafter, the value indicating the tone strength is referred to as “sound strength value 3”.

  Subsequently, the point echo diagram calculation unit 902 calculates the reciprocal of a value obtained by multiplying the square of the total length of the route indicated by the route data by 4π, and multiplies the sound intensity value 3 calculated previously by the reciprocal. The numerical value obtained by the multiplication indicates the sound intensity at the sound receiving position in consideration of all directivity, air sound absorption coefficient, reflection sound absorption coefficient, and divergence. Hereinafter, the numerical value is referred to as “attenuation sound intensity value”.

  Subsequently, the point echo diagram calculation unit 902 calculates the speed of sound using the temperature indicated by the temperature / humidity data D201. At that time, the point echo diagram calculation unit 902 uses, for example, “sound speed (m / sec) = 331 + (temperature × 0.6)” as a calculation formula of the sound speed. The point echo diagram calculation unit 902 divides the total length of the route indicated by the route data by the calculated sound velocity, thereby obtaining a time until the sound emitted at the sound source position reaches the sound receiving position (hereinafter, “arrival”). Called time).

  When the point echo diagram calculation unit 902 calculates the sound arrival time and the attenuated sound intensity value for all the route data as described above for each reference frequency point, as in the case of the sound field simulator 90, those points are used. A data group indicating a combination of numerical values is created as point echo diagram data D907.

  Subsequently, the point echo diagram calculation unit 902 creates the point echo diagram data D907 for each interpolation frequency point by the same procedure as that for the reference frequency point. However, in that case, the point echo diagram calculation unit 902 does not use the directivity data D203, the air sound absorption coefficient data D204, and the reflection sound absorption coefficient data D904, but the interpolation directivity data D206, the interpolated air sound absorption coefficient data D207, and the interpolated reflection sound absorption coefficient data. D208 is used.

  The above is the procedure for the point echo diagram calculation unit 902 to create the point echo diagram data D907. The point echo diagram data D907 for each of the reference frequency point and the interpolation frequency point created by the point echo diagram calculation unit 902 is stored in the temporary storage unit 906.

When the point echo diagram data D907 is stored in the temporary storage unit 906, the sound field simulator 10 gives the user an option regarding the time function used when the band echo diagram calculation unit 903 creates the band echo diagram data D908. , A screen prompting to select one is displayed. The sound field simulator 10 displays the following options, for example.
(A) Rectangular (no crossfade)
(B) Sine shape (c) Saw shape (d) Trapezoid

  When the user selects option (a) on the above screen, the band echo diagram calculation unit 903 focuses on the reference frequency point or the interpolation frequency point as shown in FIG. The band echo diagram data D908 is created by convolving a coefficient sequence of a time waveform including rectangular frequency components as frequency points into the corresponding point echo diagram data D907.

When the user selects the option (b) on the above screen, the band echo diagram calculation unit 903 performs the echo diagram shown by the point echo diagram data D907 regarding the reference frequency point and the interpolation frequency point with respect to the echo diagram shown in FIG. Band echo diagram data D908 relating to a frequency band having the reference frequency point or the interpolation frequency point as the center frequency point is generated by convolving each coefficient sequence of the time waveform having the frequency components shown in (n). Here, f ₁ ~f _n shown in FIG. 8 (1) ~ (n) is obtained by arranging the reference frequency points and interpolating frequency points in order from the low frequency side. This also applies to FIGS. 9 and 10 below.

The amplitude value of the frequency component included in FIG. 8 (1), that is, the time waveform convoluted to the lowest reference frequency point is given by, for example, the following function.

The amplitude values of the frequency components shown in FIGS. 8 (2) to (n-1) are given by the following function, for example.

The amplitude value of the frequency component included in FIG. 8 (n), that is, the time waveform convolved with the reference frequency point on the highest frequency side is given by, for example, the following function.

  When the user selects the option (c) on the above screen, the band echo diagram calculation unit 903 performs the echo diagram shown by the point echo diagram data D907 regarding the reference frequency point and the interpolation frequency point with respect to the echo diagram shown in FIG. Band echo diagram data D908 relating to a frequency band having the reference frequency point or the interpolation frequency point as the center frequency point is generated by convolving each coefficient sequence of the time waveform having the frequency components shown in (n).

  When the user selects the option (d) on the above screen, the band echo diagram calculation unit 903 performs the echo diagram shown by the point echo diagram data D907 regarding the reference frequency point and the interpolation frequency point with respect to the echo diagram shown in FIG. Band echo diagram data D908 relating to a frequency band having the reference frequency point or the interpolation frequency point as the center frequency point is generated by convolving each coefficient sequence of the time waveform having the frequency components shown in (n).

Here, the frequency functions shown in FIGS. 8, 9, and 10 have the following characteristics in common.
(A) When the frequency functions for all frequency bands are added, the level of the frequency component becomes “1” in all frequency bands.
(B) The level of the frequency component at each center frequency point is “1”.
(C) There is an overlap portion between adjacent frequency bands, and each frequency function crossfades in the overlap portion.

  However, in practice, the level of the frequency component after addition in all frequency ranges is slightly higher or lower than “1” and the frequency component at each center frequency point is related to the conditions (a) and (b) due to rounding errors and the like. The level may be slightly higher or lower than “1”. The band echo diagram calculation unit 903 may use a frequency function having another shape that satisfies the above conditions (a) to (c) in place of the frequency function shown in FIGS. Furthermore, for the purpose of adding a special acoustic effect, for example, a frequency function having a shape that does not satisfy the conditions (a) and (b) may be used. The band echo diagram data D908 created by the band echo diagram calculation unit 903 as described above is stored in the temporary storage unit 906.

  The adding unit 904 adds all the band echo diagram data D908 stored in the temporary storage unit 906 to create impulse response data D909. The impulse response data D909 generated by the adding unit 904 is stored in the temporary storage unit 906.

  The convolution operation unit 905 convolves the sound data D905 with the coefficient sequence indicated by the impulse response data D909, thereby providing sound field information-added sound data D910 indicating the sound to which the sound field information indicated by the impulse response data D909 is added. Is generated. When the sound data D910 with sound field information is reproduced, the sound obtained according to the sound data D905 is actually reproduced at the sound source position indicated by the sound source position data D902 in the sound field indicated by the shape data D901. In addition, the sound is similar to the sound received at the sound receiving position indicated by the sound receiving position data D903.

  Except for the case where the user selects “individually specified” among the options related to the method for determining the number of interpolations described above and selects “0” as the number of interpolations between all adjacent reference frequency points, the sound field simulator 10 generates impulse response data D909 obtained by joining band echo diagram data D908 related to a narrower frequency band by interpolation processing than in the case of the sound field simulator 90. Therefore, the gap on the frequency axis of the frequency component included in the impulse response data D909 is smaller than that in the sound field simulator 90. As a result, the sound obtained by reproducing the sound field information-added sound data D910 created by the sound field simulator 10 brings a more natural audibility.

  In addition, the sound field simulator 10 does not include adjacent bands except for the case where the user selects “rectangle (no crossfade)” from the options related to the time function used when creating the above-described band echo diagram data D908. Band echo diagram data D908 is created so that the levels of the frequency components included in echo diagram data D908 cross-fade with each other. Therefore, also in this case, the gap on the frequency axis of the frequency component included in the impulse response data D909 obtained by adding the band echo diagram data D908 is smaller than that in the sound field simulator 90. As a result, the sound obtained by reproducing the sound field information-added sound data D910 created by the sound field simulator 10 brings a more natural audibility.

  By the way, in the present embodiment, the sound field simulator 10 uses data indicating the reflection sound absorption coefficient considering the material and the incident angle as the reflection sound absorption coefficient data D904, and further uses the directivity data D203 and the air sound absorption coefficient data D204. Therefore, a more detailed simulation is performed as compared with the case of the sound field simulator 90. However, like the sound field simulator 90, the sound field simulator 10 uses the reflection sound absorption coefficient data D904 indicating the reflection sound absorption coefficient without considering the material and / or the incident angle, or the directivity data D203 and the air sound absorption coefficient data. The interpolation process and the cross-fade process described above may be performed without using both or one of D204.

  In the present embodiment, the sound field simulator 10 has been described as creating the band echo diagram data D908 by convolving a coefficient sequence of a time function with the point echo diagram data D907. The point echo diagram data D907, which is a time function, is converted into a frequency function by Fourier transform or the like, and the frequency function obtained by multiplying the frequency function by the frequency function shown in FIGS. The band echo diagram data D908 may be created by converting into a time function by conversion or the like. As described above, since the calculation by the time function has an equivalent calculation by the frequency function, a part or all of the calculation process between the time functions in the series of processes described above is replaced with an equivalent calculation process between the frequency functions. Also good.

  Further, the processing for creating the band echo diagram data D908 using the point echo diagram data D907 may be performed using an existing filter device.

[2. Second Embodiment]
Similar to the sound field simulator 10, the data processing apparatus (hereinafter referred to as “sound field simulator 30”) according to the second embodiment of the present invention uses impulse response data for adding sound field information to sound data. This is a device that generates sound data indicating a sound to which sound field information is added by performing a process of generating and convolving the generated impulse response data with the sound data. Since the sound field simulator 30 is common in many respects to the sound field simulator 10 in the first embodiment, hereinafter, the sound field simulator 30 will be described focusing on differences from the sound field simulator 10.

  FIG. 11 is a diagram illustrating a functional configuration of the sound field simulator 30. First, the procedure for creating the route data group D906 by the route calculation unit 901 and the procedure for creating the interpolation number data D205 by the interpolation number determination unit 201 in this embodiment are the same as those in the first embodiment.

  Next, regarding the interpolation processing, the processing in this embodiment is different from the processing in the first embodiment. In the sound field simulator 10 of the first embodiment, based on the interpolation number data D205, the interpolation directivity calculation unit 202, the interpolated air sound absorption rate calculation unit 203, and the interpolated reflection sound absorption rate calculation unit 204 are respectively converted into directivity data D203, air Interpolation directivity data D206, interpolated air sound absorption coefficient data D207, and interpolated reflection sound absorption coefficient data D208 are created as data for interpolating sound absorption coefficient data D204 and reflection sound absorption coefficient data D904. On the other hand, the sound field simulator 30 of the present embodiment does not include the interpolation directivity calculation unit 202, the interpolated air sound absorption rate calculation unit 203, and the interpolated reflection sound absorption rate calculation unit 204. Therefore, the interpolation directivity data D206, the interpolation The air sound absorption coefficient data D207 and the interpolated reflection sound absorption coefficient data D208 are not created. That is, the point echo diagram calculation unit 902 creates only the point echo diagram data D907 related to the reference frequency point.

  On the other hand, the sound field simulator 30 of this embodiment includes an interpolation point echo diagram calculation unit 301. The interpolation point echo diagram calculation unit 301 is a component that generates interpolation point echo diagram data D301 indicating an echo diagram related to an interpolation frequency point from point echo diagram data D907 related to a reference frequency point. Hereinafter, processing of the interpolation point echo diagram calculation unit 301 will be described.

  When the point echo diagram data D907 related to the reference frequency point is stored in the temporary storage unit 906, the interpolation point echo diagram calculation unit 301 includes the data indicating the level of sound intensity related to the same time included in the point echo diagram data D907. Is read. Subsequently, the interpolation point echo diagram calculation unit 301 calculates a spline curve that passes through all points having the numerical value indicating the sound intensity indicated by the read data as the y coordinate and the corresponding reference frequency point as the x coordinate. Subsequently, the interpolation point echo diagram calculation unit 301 is a point on the calculated spline curve, and when the x axis is logarithmically displayed, the number between x coordinates indicating each reference frequency point is a number indicated by the interpolation number data D205. The coordinates of the points to be internally divided are calculated.

  For example, the initial number indicated by the interpolation number data D205 is “3”, and the point echo diagram data D907 of each reference frequency point, that is, 31.3 Hz, 62.5 Hz, 125 Hz,. When the sound intensity levels in seconds are indicated as “0.890”, “0.931”, “0.842”,..., The interpolation point echo diagram calculation unit 301 first calculates (31.3, 0 .890), (62.5, 0.931), (125, 0.842),..., A spline curve that passes through all points indicated by the coordinates is calculated. Subsequently, the interpolation point echo diagram calculation unit 301 is a point on the calculated spline curve, and in the logarithmic display, three numerical values that divide 31.3 and 62.5 into four equal parts, that is, 37.2, 44.2 and The y coordinate of a point having 52.6 as the x coordinate is obtained. These y-coordinates indicate an estimate of the sound intensity at 0.25 milliseconds for the interpolated frequency points, i.

  Similarly, the interpolation point echo diagram calculation unit 301 calculates a sound intensity estimation value for other interpolation frequency points at 0.25 milliseconds. Furthermore, the interpolation point echo diagram calculation unit 301 similarly calculates an estimated value of sound intensity at other times for each interpolation frequency point. The interpolation point echo diagram calculation unit 301 creates, as interpolation point echo diagram data D301 for each interpolation frequency point, data indicating the estimated sound intensity at each time for each interpolation frequency point thus calculated. The above is the processing when the interpolation point echo diagram calculation unit 301 is created by the interpolation point echo diagram calculation unit 301. The interpolation point echo diagram data D301 created by the interpolation point echo diagram calculation unit 301 is stored in the temporary storage unit 906.

  When the interpolation point echo diagram data D301 is stored in the temporary storage unit 906, the band echo diagram calculation unit 903 creates the band echo diagram data D908 as in the sound field simulator 90. However, in this embodiment, the band echo diagram calculation unit 903 is created by the interpolation point echo diagram calculation unit 301 when creating the band echo diagram data D908 related to the frequency band having each interpolation frequency point as the center frequency point. Interpolated point echo diagram data D301 is used. As a result, as in the case of the first embodiment, band echo diagram data D908 is created for the frequency band having each reference frequency point and each interpolation frequency point as the center frequency point.

  A procedure for creating impulse response data D909 by the adder 904 using the band echo diagram data D908, and sound data D910 with sound field information is created by the convolution calculator 905 using the sound data D905 and the impulse response data D909. The procedure is the same as in the first embodiment.

  As described above, the impulse response data D909 generated by the sound field simulator 30 also has a narrower frequency band than the case of the sound field simulator 90, similarly to the impulse response data D909 generated by the sound field simulator 10. Is the impulse response data D909 obtained by joining the band echo diagram data D908 with respect to. Therefore, the gap on the frequency axis of the frequency component included in the impulse response data D909 is smaller than that in the sound field simulator 90. As a result, the sound obtained by reproducing the sound field information-added sound data D910 created by the sound field simulator 10 brings a more natural audibility.

It is the figure which showed the hardware constitutions of the sound field simulator concerning embodiment of this invention. It is the block diagram which showed the function structure of the sound field simulator concerning 1st Embodiment of this invention. It is the figure which showed the structure of the shape data concerning embodiment of this invention. It is the figure which showed the structure of the reflection sound absorption coefficient data concerning embodiment of this invention. It is the figure which showed the structure of the interpolation number template data group concerning embodiment of this invention. It is the figure which showed the structure of the directivity data concerning embodiment of this invention. It is the figure which showed the structure of the air sound absorption coefficient data concerning embodiment of this invention. It is the figure which showed the frequency component of the time waveform used when creating the band echo diagram data concerning embodiment of this invention. It is the figure which showed the frequency component of the time waveform used when creating the band echo diagram data concerning embodiment of this invention. It is the figure which showed the frequency component of the time waveform used when creating the band echo diagram data concerning embodiment of this invention. It is the block diagram which showed the function structure of the sound field simulator concerning 2nd Embodiment of this invention. It is the figure which showed the function structure of the sound field simulator concerning a prior art. It is the figure which illustrated the echo diagram data concerning a prior art. It is the figure which showed the frequency component of the time waveform used when creating the band echo diagram data concerning a prior art.

Explanation of symbols

  10.30.90 ... Sound field simulator, 101 ... CPU, 102 ... ROM, 103 ... RAM, 104 ... HD, 105 ... Display, 106 ... Operation unit, 107 ... Sound system, 108 ... Input / output I / F, 201 ... Interpolation number determination unit, 202 ... Interpolation directivity calculation unit, 203 ... Interpolation air sound absorption rate calculation unit, 204 ... Interpolation reflection sound absorption rate calculation unit, 301 ... Interpolation point echo diagram calculation unit, 900 ... Storage unit, 901 ... Path calculation unit 902: Point echo diagram calculation unit, 903: Band echo diagram calculation unit, 904 ... Addition unit, 905 ... Convolution calculation unit, 906 ... Temporary storage unit, D200 ... Sound source name data, D201 ... Temperature / humidity data, D202 ... Number of interpolations Template data group, D203 ... directivity data, D204 ... air sound absorption coefficient data, D205 ... complement Numerical data, D206 ... Interpolation directivity data, D207 ... Interpolated air sound absorption coefficient data, D208 ... Interpolation reflection sound absorption coefficient data, D301 ... Interpolation point echo diagram data, D901 ... Shape data, D902 ... Sound source position data, D903 ... Sound reception position Data, D904: Reflected sound absorption coefficient data, D905 ... Sound data, D906 ... Path data group, D907 ... Point echo diagram data, D908 ... Band echo diagram data, D909 ... Impulse response data, D910 ... Sound data with sound field information.

Claims (3)

Using the feature data indicating the characteristics of the sound field, the sound source position data indicating the sound source position in the sound field, and the sound receiving position data indicating the sound receiving position in the sound field, each sound of a plurality of reference frequencies is converted into the sound source position. Calculating for each of the plurality of reference frequencies an echo diagram representing a temporal change in the sound intensity level of the sound that reaches the sound receiving position when generated in
The number of interpolation frequencies between the plurality of reference frequencies is a predetermined fixed number, a specified number, a number associated with the selected music genre, directivity of the sound source in the sound field, the reference frequency The number set based on the air sound absorption rate and the reflection sound absorption rate related to the reference frequency, the frequency component included in the time waveform of the sound before the reverberation sound is added, and the frequency band between the adjacent reference frequencies is calculated. A process determined to be one of the numbers set according to the magnitude of the amplitude value of the included frequency component,
Directivity data in which the plurality of reference frequencies are associated with sound intensity levels at predetermined positions when a sound of the reference frequency is generated at the sound source position is acquired, and the frequency is an x coordinate on a logarithmic axis. In the coordinate having the sound intensity level as the y coordinate, the reference frequency in the acquired directivity data is set as the x coordinate, the sound intensity level in the directivity data is set as the y coordinate, and the plurality of reference frequencies and the respective references at the same time. A line passing through the coordinates corresponding to the sound intensity level of the frequency is calculated, the line between the coordinates corresponding to a plurality of reference frequencies is equally divided based on the determined number of interpolation frequencies, and the coordinates of the dividing positions are divided. Calculating an echo diagram at each of one or more interpolation frequencies corresponding to
For each of the plurality of reference frequencies and the interpolation frequency, a coefficient sequence representing a time waveform having a positive frequency component only in a pass band including the reference frequency or the interpolation frequency as a center frequency and including the center frequency, The process of convolution into the echo diagram corresponding to the frequency and the interpolation frequency,
Generating impulse response waveform data representing transfer characteristics of sound in the sound field by adding the convolved echo diagrams for each of the plurality of reference frequencies and the interpolation frequency Method. The feature data includes information indicating a sound absorption coefficient of the boundary surface of the sound field at each of the plurality of reference frequencies,
The impulse response data generation method includes:
Interpolating the sound absorption rate at each of the plurality of reference frequencies to calculate a sound absorption rate at each of the one or more interpolation frequencies between the plurality of reference frequencies;
When the sound at each of the one or more interpolation frequencies is generated at the sound source position using the characteristic data, the sound source position data, the sound receiving position data, and the sound absorption rate at each of the one or more interpolation frequencies. By calculating, for each of the one or more interpolation frequencies, an echo diagram representing a temporal change in the level of the sound reaching the sound receiving position, at each of the one or more interpolation frequencies between the plurality of reference frequencies. 2. The impulse response data generation method according to claim 1, wherein an echo diagram is calculated. Using the feature data indicating the characteristics of the sound field, the sound source position data indicating the sound source position in the sound field, and the sound receiving position data indicating the sound receiving position in the sound field, each sound of a plurality of reference frequencies is converted into the sound source position. Calculating for each of the plurality of reference frequencies an echo diagram representing a temporal change in the sound intensity level of the sound that reaches the sound receiving position when generated in
The number of interpolation frequencies between the plurality of reference frequencies is a predetermined fixed number, a specified number, a number associated with the selected music genre, directivity of the sound source in the sound field, the reference frequency The number set based on the air sound absorption rate and the reflection sound absorption rate related to the reference frequency, the frequency component included in the time waveform of the sound before the reverberation sound is added, and the frequency band between the adjacent reference frequencies is calculated. A process determined to be one of the numbers set according to the magnitude of the amplitude value of the included frequency component,
Directivity data in which the plurality of reference frequencies are associated with sound intensity levels at predetermined positions when a sound of the reference frequency is generated at the sound source position is acquired, and the frequency is an x coordinate on a logarithmic axis. In the coordinate having the sound intensity level as the y coordinate, the reference frequency in the acquired directivity data is set as the x coordinate, the sound intensity level in the directivity data is set as the y coordinate, and the plurality of reference frequencies and the respective references at the same time. A line passing through the coordinates corresponding to the sound intensity level of the frequency is calculated, the line between the coordinates corresponding to a plurality of reference frequencies is equally divided based on the determined number of interpolation frequencies, and the coordinates of the dividing positions are divided. Calculating an echo diagram at each of one or more interpolation frequencies corresponding to
For each of the plurality of reference frequencies and the one or more interpolation frequencies, a frequency component that attenuates by drawing a slope toward the low frequency side and the high frequency side of each center frequency with the reference frequency or the interpolation frequency as the center frequency. A process of convolving a coefficient sequence representing a time waveform having a reference frequency and an echo diagram corresponding to the interpolation frequency,
Generating impulse response waveform data representing sound transfer characteristics in the sound field by adding the convolved echo diagrams for each of the plurality of reference frequencies and the one or more interpolated frequencies. Response data generation method.

Images