JP2014045282A - Reverberation adding device, reverberation adding program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To generate reverberation components, on which the direction information of a sound source in a reproduced acoustic signal is reflected, from a reverberation model obtained by radiating a sound from a sound source at one specific point.SOLUTION: A reverberation adding device 1 includes at least three speakers SP, a storage unit 20 for storing a reverberation model of each speaker SP obtained by radiating a sound from one sound source, a reverberation generation unit 30 for generating the reverberation components of each speaker by convoluting the reverberation model into an acoustic signal, and a reverberation reconfiguration unit 40 for reconfiguring the reverberation components of each speaker, depending on the angle formed by the sound source direction in the reverberation model and the sound source direction in the acoustic signal.

Description

  The present invention relates to a reverberation adding apparatus and a reverberation adding program.

  In multi-channel sound production for TV programs, music, movies, etc., a technique for adding reverberation is used to give a sense of sound spread and presence. IIR (Infinite Impulse Response) filter type that simulates reverberation in space, and FIR (Finite Impulse Response) filter type that generates reverberation using impulse response measured in real space Various reverberation adding devices have been developed.

  By the way, the multi-channel sound system such as the 5.1ch surround system and the 22.2 multi-channel system for Super Hi-Vision has a feature that the sound source can be localized in various directions on a plane or in space (for example, Non-patent document 1). In general, in the process of producing these contents, a reverberation adding device is often used to give a rich sense of presence and spatial spread. However, when the sound source direction changes in real space, each reflection follows it. Ideally, the reverberation component reproduced from each speaker changes following the sound source direction even in the multi-channel sound system so that the reflected sound from the surface changes. In recent years, reverberation adding apparatuses for multi-channel sound have also been developed (for example, see Non-Patent Document 2).

Akio Ando “Highly Realistic Acoustic Technology and Its Theory” IEICE Fundamental Review Vol.3 No.4 pp.33-46 Apr 2010 W.woszczyk, et al., “Space Builder: An impulse response-based tool for immersive 22.2 channel ambience design,” AES 40th international conference, Tokyo, Japan (2010)

  However, as in Non-Patent Document 1, in the conventional method of generating reverberation by simulation or actual measurement, it is common to use an impulse response under a condition in which sound is emitted from a certain sound source position. Like the channel sound reproduction method, it is not suitable for a method in which the sound source direction (sound arrival direction) varies in various ways. In addition, it is theoretically possible to measure an infinite number of different impulse response sets corresponding to each sound source direction, and generate and add reverberation components to each. However, from the viewpoint of measurement, the complexity of the filter bank However, this is not a realistic configuration.

  Also, the reverberation adding device for multi-channel sound described in Non-Patent Document 2 is not configured to use the direction information of the sound source as in the conventional method. is there.

  Accordingly, an object of the present invention made in view of such a point is that an acoustic signal to be reproduced from a reverberation model (a set of impulse responses) obtained by radiating sound from a sound source at a specific location by simulation or actual measurement. An object of the present invention is to provide a reverberation adding apparatus and a reverberation adding program capable of generating a reverberation component reflecting direction information of a sound source.

  In order to solve the above-described problems, a reverberation adding device according to the present invention includes at least three speakers, a storage unit that stores a reverberation model of each speaker obtained by radiating sound from one sound source, and an acoustic signal. The reverberation component of each speaker is convoluted with the reverberation generator for generating the reverberation component of each speaker, the sound source direction in the reverberation model, and the angle formed by the sound source direction in the acoustic signal. And a reverberation reconstructing unit for reconstructing.

  Moreover, it is preferable that the said reverberation reconstruction part changes distribution of the said reverberation component with respect to each said speaker according to the said angle as reconstruction of the said reverberation component.

  The reverberation reconstruction unit calculates a weight component for each adjacent speaker by applying an amplitude panning method to the adjacent speakers, and adds the reverberation component obtained by multiplying the weight components between the adjacent speakers. By doing so, it is preferable to reconstruct the reverberation component.

  In addition, a reverberation adding program according to the present invention includes a reverberation adding apparatus including at least three speakers and a storage unit that stores a reverberation model of each speaker obtained by radiating sound from one sound source. The reverberation component of each speaker is reconstructed according to the step of generating the reverberation component of each speaker by convolving the reverberation model, and the angle formed by the sound source direction in the reverberation model and the sound source direction in the acoustic signal. Step.

  According to the reverberation adding apparatus and the reverberation adding program according to the present invention, in a sound signal to be reproduced from a reverberation model (a set of impulse responses) obtained by radiating sound from a sound source at a specific location by simulation or actual measurement. It is possible to generate a reverberation component reflecting the direction information of the sound source.

It is a figure which shows the functional block of the reverberation addition apparatus 1 which concerns on one Embodiment of this invention. It is a figure which shows an example of the set of measurement space and the obtained impulse response. It is a figure for demonstrating redistribution of the reverberation component by the direction of a virtual sound source. It is a figure which shows an example of the reconfiguration | reconstruction of the reverberation component according to the angle of the virtual sound source in FIG. It is a figure which shows the outline | summary of an amplitude panning method. It is a figure which shows an example of the reconstruction of the reverberation component adjusted based on the amplitude panning method.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a diagram showing functional blocks of a reverberation adding apparatus 1 according to an embodiment of the present invention. The reverberation adding apparatus 1 includes a decoder 10, an impulse response storage unit 20, a reverberation generation unit 30, a reverberation reconstruction unit 40, an amplifier 50, and a speaker SP. Here, the speaker SP includes at least three speakers, and in the following description, the nth speaker is appropriately referred to as a speaker SPn.

  The decoder 10 decodes the input acoustic signal and outputs the decoded acoustic signal to the reverberation generation unit 30. Here, various multi-channel sound systems such as 5.1 channel sound used in home theaters and digital broadcasting and 22.2 channel sound that is super high-definition sound can be used as the sound signal. The acoustic signal in the present embodiment also includes direction information that represents the direction of the sound source (virtual sound source) in the acoustic signal. The direction information is, for example, an identifier of each channel (speaker) in the multi-channel sound system, an angle formed by the virtual sound source with respect to the front of the listener, and the direction information is transmitted to the decoder 10 as a signal integrated with the sound signal. The direction information can be input to the decoder 10 as a signal independent of the acoustic signal.

  The impulse response storage unit 20 stores a reverberation model (a set of impulse responses) of each speaker SP obtained by radiating sound from one sound source. A set of impulse responses can be generated by measurement in real space or by computer simulation. In the following description, it is assumed that a reverberation model that is a set of impulse responses is X (t), and an impulse response of each speaker SPn is represented by Xn (t).

FIG. 2 is a diagram illustrating an example of a measurement space and a set of obtained impulse responses. In the rectangular measurement space shown in FIG. 2A, when a signal for impulse response measurement such as a TSP (Time Stretched Pulse) signal is radiated from the sound source SP _source , the impulse response X1 ( t) to X4 (t) are recorded. As shown in FIG. 2B, the characteristics of the impulse responses X1 (t) to X4 (t) of the microphones MC1 to MC4 have different characteristics depending on the sound source SP _source , the difference in distance from the wall surface, and the like. Here, the number and arrangement of the microphones MC used for the measurement of the reverberation model correspond to the number and arrangement of the speakers SP of the reverberation adding apparatus 1. In the case of FIG. 2, impulse responses X1 (t) to X4 (t) represent impulse responses that generate reverberation components to be reproduced from the speakers SP1 to SP4 of the reverberation adding device 1 including the four speakers SP1 to SP4. is there.

  The reverberation generating unit 30 convolves the acoustic signal from the decoder 10 with the reverberation model X (t) stored in the impulse response storage unit 20, and the sound source in one specific direction (the sound source direction when the reverberation model X (t) is measured). A reverberation component corresponding to is generated. In the following description, a set of reverberation components is R (t), and a reverberation component of each speaker SPn is represented by Rn (t). For example, the reverberation component R generated when the reverberation model X (t) = (X1 (t), X2 (t), X3 (t), X4 (t)) measured in FIG. (T) shall be expressed as R (t) = (R1 (t), R2 (t), R3 (t), R4 (t)).

  The reverberation reconstruction unit 40 reconstructs the reverberation component Rn (t) of each speaker SPn according to the angle formed by the sound source direction in the reverberation model and the sound source direction in the acoustic signal. That is, the reverberation reconstructing unit 40 has a reverberation component R (t) corresponding to a sound source in one specific direction (a sound source direction at the time of reverberation model X (t) measurement) of the virtual sound source of the acoustic signal currently being reproduced. The distribution of each reverberation component Rn (t) to each speaker SPn is changed so that the reverberation component R ′ (t) corresponds to the direction.

  FIG. 3 is a diagram for explaining redistribution of reverberation components depending on the direction of the virtual sound source. Here, it is assumed that the sound source direction at the time of reverberation model X (t) measurement is the direction of MC1 (corresponding to the direction of SP1 (front of the listener) in FIG. 3) as shown in FIG. The reverberation reconstruction unit 40 redistributes the reverberation component R (t) according to the angle θ of the virtual sound source with respect to the front of the listening position. For example, when the virtual sound source is in front of the listener (θ = 0 degrees), the corresponding reverberation components R1 (t) to R4 (t) may be reproduced as they are from the speakers SP1 to SP4. In this case, R ′ (t) = R (t), and the reverberation reconstruction unit 40 outputs the reverberation component R (t) generated by the reverberation generation unit 30 to the amplifier 50 as it is.

On the other hand, when the direction of the virtual sound source is a predetermined angle or more with respect to the front of the listener, the reverberation reconstruction unit 40 changes the distribution of the reverberation components R1 (t) to R4 (t) to the speakers SP1 to SP4. can do. In the case of FIG. 3, if the angle between adjacent speakers SP is 2θ ₀ , then 2θ ₀ = 90 degrees, and 45 degrees that is θ ₀ is an intermediate angle between adjacent speakers SP. For example, when the direction of the virtual sound source is 90 degrees (> θ ₀ ) with respect to the front of the listener, the reverberation reconstruction unit 40 uses the reverberation component R ′ (t) distributed to the speakers SP1 to SP4. , R ′ (t) = (R4 (t), R1 (t), R2 (t), R3 (t)).

  FIG. 4 is a diagram illustrating an example of reconfiguration of reverberation components according to the angle θ of the virtual sound source in FIG. For example, when the angle θ is −45 degrees <θ ≦ 45 degrees, the reverberation reconstruction unit 40 outputs the reverberation component R (t) generated by the reverberation generation unit 30 to the amplifier 50 as R ′ (t) as it is. . For example, when the angle θ is 135 degrees <θ ≦ 225 degrees, the reverberation reconstruction unit 40 rotates the reverberation component R (t) generated by the reverberation generation unit 30, and R ′ (t) = (R3 (T), R4 (t), R1 (t), R2 (t)) are output to the amplifier 50. Thus, the reverberation component R (t) corresponding to the sound source in one specific direction (the sound source direction at the time of the reverberation model X (t) measurement) is reverberated according to the direction of the virtual sound source of the acoustic signal currently being reproduced. By reconfiguring as the component R ′ (t), it is possible to add a reverberation component corresponding to the sound signal being reproduced based on a set of reverberation models.

  The reverberation reconstruction unit 40 can reconstruct the reverberation component R (t) based on the amplitude panning method (see Non-Patent Document 1). In the following description, it is assumed that the reverberation component of each speaker SPn by the amplitude panning method is represented by R′n (t). That is, R ′ (t), which is a reverberation component after reconstruction, is represented by R ′ (t) = (R′1 (t), R′2 (t), R′3 (t), R′4 (t )).

FIG. 5 is a diagram showing an outline of the amplitude panning method. Consider a case where an acoustic signal is distributed to left and right speakers SP _L and SP _R by weighting factors W _L and W _R. When the angle of each speaker SP _L and SP _R with respect to the front of the listener is θ ₀ and the angle from the left speaker SP _L (ie, the sound source direction at the time of the reverberation model measurement in FIG. 2A) to the virtual sound source is θ. In the tangent rule in the amplitude panning method, the weighting factors W _L and W _R for localizing the sound at the angle θ are expressed by Expression (1).

Here, when 1 the sum of the weighting factors _{W L} and _{W R,} the denominator of the right side becomes 1 of the formula (1), the weight coefficient _{W L} and _{W R} formula (2) is derived.

When the amplitude panning method is applied to each of the speakers SP1 to SP4 shown in FIG. 3, the weighting coefficient in Expression (2) may be applied between the speakers adjacent in the reproduction space. For example, in FIG. 3, the speaker SP1 and the speaker SP4 are adjacent to each other. When the speaker SP1 corresponds to the speaker SP _R and the speaker SP4 corresponds to the speaker SP _L , the reverberation component R1 ′ ( t) is expressed by equation (3).

  Similarly, by applying the amplitude panning method between the adjacent speakers SP1 and SP2, in FIG. The reverberation components R2 ′ (t), R3 ′ (t), and R4 ′ (t) to be reproduced are expressed by the following equation (4).

For example, when the direction of the virtual sound source is 45 degrees (= θ ₀ ) with respect to the front of the listener, the reverberation reconstruction unit 40 uses the reverberation component R ′ (t) distributed to the speakers SP1 to SP4. And reconstructing based on equations (3) and (4). That is, the reverberation components R1 ′ (t) to R4 ′ (t) to be reproduced from the speakers SP1 to SP4 after the reverberation reconstruction are expressed by Expression (5).

  The reverberation reconstruction unit 40 can reconstruct the reverberation component by appropriately combining a method of changing the distribution of the reverberation component to the speakers SP1 to SP4 and a method of applying the amplitude panning method between adjacent speakers.

For example, when the angle θ exceeds the intermediate angle θ ₀ between the adjacent speakers SP, the distribution of the reverberation component R (t) is changed, and the reverberation component R ′ (t) is regenerated based on the amplitude panning method. Can be configured. In this case, for example, when the direction of the virtual sound source is 135 degrees with respect to the front of the listener, the reverberation reconstruction unit 40 determines the reverberation component R ′ (t) distributed to the speakers SP1 to SP4 in FIG. After changing according to the table, the reverberation component R ′ (t) can be reconstructed based on the amplitude panning method. Specifically, the reverberation reconstruction unit 40 determines the reverberation components R ′ (t) distributed to the speakers SP1 to SP4 as R ′ (t) = (R4 (t), R1 (t), R2 (t ), R3 (t)), and the reverberation components R1 ′ (t) to R4 ′ (t) can be reconstructed as shown in equation (6) based on the amplitude panning method shown in equation (2). .

  Further, for example, when the angle θ stays in the range between the adjacent speakers SP, the reverberation component R ′ (t) is reconstructed based on the amplitude panning method, and when the angle θ exceeds the adjacent speakers SP, The distribution of the reverberation component R ′ (t) adjusted based on the amplitude panning method can be changed. FIG. 6 is a diagram illustrating an example of reconstructing reverberation components adjusted based on the amplitude panning method. In this case, for example, when the virtual sound source is in front of the listener (θ = 0 degree), the corresponding reverberation components R1 (t) to R4 (t) may be reproduced as they are from the speakers SP1 to SP4. . For example, when the angle θ is 0 degree <θ <90 degrees, the reverberation reconstruction unit 40 applies the amplitude panning method to the reverberation component R (t) generated by the reverberation generation unit 30, and R ′ (t) = (R′1 (t), R′2 (t), R′3 (t), R′4 (t)). When the angle θ is 90 degrees, the reverberation reconstruction unit 40 rotates the reverberation component R ′ (t) distributed to the speakers SP1 to SP4, and R ′ (t) = (R4 (t), R1 (t), R2 (t), and R3 (t)) are output to the amplifier 50. Further, when the angle θ is 180 degrees <θ <270 degrees, the reverberation reconstruction unit 40 applies the amplitude panning method to the reverberation component R (t) generated by the reverberation generation unit 30, and further, according to FIG. The distribution of the reverberation component R ′ (t) adjusted based on the panning method is changed, and R ′ (t) = (R′3 (t), R′4 (t), R′1 (t), R ′. 2 (t)) and output to the amplifier 50.

  The reverberation reconstruction unit 40 adds the reconstructed reverberation component R ′ (t) to the acoustic signal and outputs it to the amplifier 50. The amplifier 50 amplifies the acoustic signal to which the reverberation component R ′ (t) is added, and the speaker SP reproduces the acoustic signal to which the reverberation component R ′ (t) is added as sound.

  Thus, according to the present embodiment, the impulse response storage unit 20 stores the reverberation model of each speaker obtained by radiating sound from one sound source, and the reverberation generation unit 30 adds the reverberation model to the acoustic signal. The reverberation component of each convolution speaker is generated, and the reverberation reconstruction unit 40 reconstructs the reverberation component of each speaker SP according to the angle formed by the sound source direction in the reverberation model and the sound source direction in the acoustic signal. As a result, a reverberation component reflecting the direction information of the sound source in the sound signal to be reproduced is generated from a reverberation model (a set of impulse responses) obtained by radiating sound from one specific sound source by simulation or actual measurement. It becomes possible to do.

  Moreover, the reverberation reconstruction unit 40 changes the distribution of the reverberation component to each speaker SP in accordance with the angle formed by the sound source direction in the reverberation model and the sound source direction in the acoustic signal as reverberation component reconstruction. Thereby, it is possible to generate a reverberation component reflecting the direction information of the sound source in the acoustic signal to be reproduced from the reverberation model without performing complicated calculation processing.

  Further, the reverberation reconstruction unit 40 calculates the weight component for each adjacent speaker SP by applying the amplitude panning method to the adjacent speakers SP, and adds the reverberation component multiplied by the weight component between the adjacent speakers SP. Thus, the reverberation component is reconstructed. This makes it possible to generate a reverberation component reflecting the direction information of the sound source in the acoustic signal to be reproduced from the reverberation model when there is a virtual sound source in a place where there is no physical speaker.

  Although the present invention has been described based on the drawings and examples, it should be noted that those skilled in the art can easily make various modifications and corrections based on the present disclosure. Therefore, it should be noted that these variations and modifications are included in the scope of the present invention. For example, the functions included in each component, each step, etc. can be rearranged so that there is no logical contradiction, and multiple components, steps, etc. can be combined or divided into one It is.

For example, in the arrangement of the loudspeakers SP shown in FIG. 3, all the loudspeakers SP are arranged at equal intervals so that θ ₀ that is an intermediate angle between adjacent speakers is matched. As described above, even when the angles formed by the adjacent reproduction speakers are different, the weighting coefficient may be calculated by applying the amplitude panning formula to each adjacent speaker. Moreover, although the example of the reverberation addition in four speakers was described in the said embodiment, this invention is applicable regardless of the number of some if it is three or more speakers. Furthermore, in the above embodiment, reverberation generation on a two-dimensional plane is assumed, but even when two-dimensional planes are superimposed three-dimensionally, by performing the same processing on each two-dimensional plane, the third order Needless to say, expansion to the original space is possible.

  It should be understood that the present invention can also be realized as a program that causes a processor included in the reverberation adding apparatus 1 to execute an equivalent process (step), and these are also included in the scope of the present invention. For example, the reverberation adding apparatus 1 can store a program describing the processing contents for realizing each function in a storage unit, and read and execute the program by a central processing unit (CPU).

  According to the present invention, in a multi-channel acoustic system such as the 5.1ch surround system or 22.2 multi-channel system for Super Hi-Vision, a reverberation model (impulse response From the set, there is a usefulness that it is possible to generate a reverberation component reflecting the direction information of the sound source in the acoustic signal to be reproduced.

DESCRIPTION OF SYMBOLS 1 Reverberation addition apparatus 10 Decoder 20 Impulse response memory | storage part (memory | storage part)
30 Reverberation generator 40 Reverberation reconstruction unit 50 Amplifier SP Speaker MC Microphone

Claims (4)

At least three speakers;
A storage unit for storing a reverberation model of each speaker obtained by emitting sound from one sound source;
A reverberation generator that convolves the reverberation model with an acoustic signal to generate a reverberation component of each speaker;
A reverberation adding device comprising: a reverberation reconfiguring unit that reconfigures the reverberation component of each speaker according to an angle formed by a sound source direction in the reverberation model and a sound source direction in the acoustic signal.   The reverberation adding device according to claim 1, wherein the reverberation reconstruction unit changes distribution of the reverberation component to each speaker according to the angle as reconstruction of the reverberation component.   The reverberation reconstruction unit calculates a weight component for each adjacent speaker by applying an amplitude panning method to the adjacent speakers, and adds the reverberation component multiplied by the weight component between the adjacent speakers. The reverberation adding apparatus according to claim 1, wherein the reverberation component is reconfigured by the following. A reverberation adding device comprising at least three speakers and a storage unit for storing a reverberation model of each speaker obtained by radiating sound from one sound source,
Convolving the reverberation model with an acoustic signal to generate a reverberation component of each speaker;
A reverberation adding program for executing the step of reconfiguring the reverberation component of each speaker according to an angle formed by a sound source direction in the reverberation model and a sound source direction in the acoustic signal.

Images