JP2005196086A - Method and device for reverberation processing - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method and device for reverberation processing that can reduce reverberation processing quantity, simplify electric circuit design, and generate preferred reverberations. <P>SOLUTION: The reverberation processing method includes a step 201 of arranging a filter module 225 of high quality and a filter module 250 of low quality, a step 203 of inputting some sound quality signal to the filter module 225 of high quality and generating reverberations of high quality, within a finite limited time; a step 205 of inputting the sound quality signal to the filter module 250 of low quality and generating reverberations of low quality; a step 207 of delaying reverberations generated by the filter module 250 of low quality; and a step 209 of putting together the reverberations of the high quality and reverberations of low quality that the filter module 225 of high quality and the filter module 250 of low quality output. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a reverberation processing method and a reverberation processing device, and particularly provides a reverberation processing device formed by combining a high-quality filtering module and a low-quality filtering module, and obtains them within the first finite time. The present invention relates to a reverberation processing method and a reverberation processing apparatus characterized by using what is obtained by the high-quality filtering module as reverberation.

  At present, the virtual world created by the virtual sound playback device will entertain the user with a new boundary, and the user will be able to enjoy the feeling that he / she has dealt with the actual boundary through the sound he / she listens to. Sound effects equipment designed to generate sound effects will be designed to be more three-dimensional and immersive, thereby making it easier for users to obtain audio means that are increasingly more realistic. It is like that.

  Among them, the most common method of processing so as to have a three-dimensional effect and a sense of presence is to use artificially created reverberation. Generally speaking, so-called reverberation means that the original sound is predetermined. After being emitted to the environment, echo and resonance sound generated in the environment gradually disappears over time, and this increasingly disappearing sound phenomenon is called reverberation. The reverberation generation method is a filter design method, and different types of reverberation filtering devices can be designed according to different reverberation characteristics and delay times. When performing the convolution operation, reverberation can be generated. Among them, there are roughly two types of reverberation filtering device design methods, namely, a finite impulse response filter (FIR filter) and an infinite impulse response filter (IIR filter).

  Among them, the finite impulse response filter is obtained by directly measuring the reverberation of the reverberation in the environment, and the pulse response by this type of reverberation generation method has a limit in length. Although the most natural effect can be obtained, the drawback is that the computational complexity is relatively high, just as shown in FIG. 1, where the input sound quality signal is a finite impulse response filter. It is a reaction produced when 50 is passed. Among them, the delay unit 10 is used to simulate the special effect of early reflections in space, and in the case of 48K sound sampling every second, in order to obtain a reverberation of 3 seconds, It is necessary to process the convolution operation of about a point and sample. Therefore, in this drawing, the finite impulse response filter 50 can be sampled only when the longer one row of delay units 10 is arranged, which not only complicates the design of the electric circuit, but also the processing. It can make the speed quite slow.

  The latter infinite impulse response filter is a simple all-pass filter, thereby generating reverberant speech having an infinite length. It is simple to calculate, but its reverberation effect is unnatural, just as shown in FIG. 2, in which the infinite impulse response filter 100 uses a power amplifier 75 to provide feedback. And reverberant speech can be continued indefinitely. As can be seen from this drawing, the infinite impulse response filter 100 is easier to design and calculate the electric circuit than the finite impulse response filter 50 described above, but the reverberation signal generated by this type of method is large. Therefore, the reverberation that can be heard from the two left and right ears becomes more unnatural.

  As can be seen from the above description of the prior art, the reverberation that can be generated by the filters used in the prior art is clearly flawed and the designer uses a finite impulse response filter 50 where the quality of the sound effect is more favorable. It is necessary to produce a filter with a higher cost and this higher complexity, and if the electrical circuit uses a simpler infinite impulse response filter 100, the output of the reverberation The sound effect was clearly unnatural.

  For this reason, the present invention is set so that the reverberation generated in a certain finite time using the output sound quality signal is generated by a high-quality filtering module, so that a preferable reverberation is generated within a predetermined time. Providing an audio playback device that can be generated is a major problem to be solved.

  In order to achieve the above object, the present invention provides a kind of reverberation processing method and reverberation processing device, in which a human being has a sound effect that can be discerned in a certain space within a certain finite time. As a key point, the method and apparatus are designed. The method mainly includes a step of arranging a high-quality filtering module and a low-quality filtering module, and the sound quality signal as the high-quality signal. Generating a high-quality reverberation in a finite time by inputting to the filtering module; and inputting the sound quality signal to the filtering module of infinite length to input a low-quality filter of infinite length. Step 3 of generating reverberation, Step 4 of delaying the reverberation generated by the low quality filtering module, the high quality filtering module and the low quality filtering module. Wherein said high-quality reverberation for outputting a step five for combining the low-quality reverberation.

(Function)
As a result, since the output sound quality signal uses the one generated by the high-quality filtering module as the reverberation generated within a certain finite time, the user's listening within the certain finite time. As the reverberation to be performed becomes high quality, the number of reverberation processes can be reduced at other extra times, so that the processing speed can be improved.

  In order to further elucidate the features and technical contents of the present invention, the preferred embodiments of the present invention will be described in detail and specifically with reference to the accompanying drawings. Needless to say, such structures are merely possible embodiments of the present invention and do not narrowly define the scope of the present invention.

  The present invention is a reverberation processing method and a reverberation processing apparatus that combine the common excellent points of a high-quality filtering module and a low-quality filtering module, and emphasizes both the sound effect and the processing speed. It is. FIG. 3 is a simple block diagram showing the configuration of the reverberation processing apparatus according to the present invention. When the common input terminal 200 receives a sound quality signal, the sound quality signal is converted into a high-quality filtering module 225, respectively. Then, by inputting the low quality filter module 250, the sound quality signal is subjected to reverberation processing in the next stage. In this embodiment, a finite impulse response filter is used as the high quality filtering module 225 and an infinite impulse response filter is used as the low quality filtering module 250. Needless to say, the present invention is not limited to the form of the present embodiment.

  According to the latest report, it is reported that the ability to discriminate hearing in a certain space exists only within a finite time. If that is the key point, the finite second son in the experiment is about 50 milliseconds, that is, when a person enters a certain space, the reverberation in the space can be identified between the books. It is only about 50 ms when entering the first time, so the present invention processes the sound quality signal of the previous 50 ms by the high quality filtering module 255, and the high quality filtering module 255 has a large amount of calculation. However, since only the sound quality signal of about 50 ms is processed, the number of points in which the convolution calculation is used does not increase, so that a complicated electric circuit can be simplified and the reverberation after 50 ms can be simplified. It is processed by a low-quality filtering module 250 made of an electric circuit, and the processing effect is not as favorable as the above-described high-quality filtering module 255. Since the good reverberation at that time cannot already be distinguished, the good reverberation is not substantially adversely affected, and the amount of calculation is relatively reduced, so that the execution speed is accelerated by a considerable amount. Since reverberation of time can be generated and the number of points calculated by the high-quality filtering module 225 is not large, the data obtained by the high-quality filtering module 225 can be calculated using a large amount of memory. Since it is not necessary to memorize | store, memory consumption can be reduced significantly.

  Also, as shown in FIG. 3, after the sound quality signal is processed by the high quality filtering module 225, a reverberation signal with a limited length is generated, and the high quality reverberation effect is produced within a certain finite time T. And then through an adder 280 with a reverberation signal with a limit on the length, then let the user listen to the high quality reverberation directly, and the user can quickly limit the length. A certain high quality reverberation signal is heard, thereby distinguishing the generated echoes and resonances in space (environment).

  After the sound quality signal is processed by the low quality filtering module 250, the output reverberation becomes an infinite reverberation signal, and the user can listen to the low quality reverberation at a predetermined time. The infinite length reverberation signal output from the filtering module 250 is first input to the delay unit 275 connected thereafter, and the infinite length reverberation signal is delayed by a predetermined delay time K before being supplied to the adder 280. Since it is transmitted, it is combined with a reverberation signal having a limit on the length, and then the user listens, so that the user listens to the infinite length low quality reverberation signal within a first time. However, the delay unit 275 is not limited to being connected to the front or rear of the low quality filtering module 250 and its main purpose is to increase the It is for causing hear the reverberation generated by the quality of the filtered module 225, if such goals, it may employ any means or any connection method as the delay unit 275 of course.

  FIG. 4 is a curve diagram showing the relationship between the signal of the present invention and time. As can be seen from the drawing, the first output reverberation signal is a high-quality filtering module 225 on the time axis. Since the reverberation signal is a type of reverberation signal with a limited length, its output time is limited to a finite time T, and in the first case a low quality filter is used. Since the reverberation signal of infinite length output from the wave module 250 is delayed by the delay unit 275, after a predetermined delay time K has elapsed (in this time, a delay time of about 25 ms to 45 ms can be adopted as the delay time K). Are output with the reverberation signal having a limit on the length, in which a stacking time (T-K) is adopted as the stacking time, and this type of design has a limit on the length. It is a design for preventing the signal from suddenly cutting and generating a large head feeling, in other words, when the reverberation signal with a limited length approaches the end of the finite time T, the infinite length The reverberation signal is already stacked with a reverberation signal whose length is limited, so when the finite time T has passed, the user suddenly cuts off the reverberation signal whose length is limited and feels a large drop. Never generated.

  FIG. 5 is a flowchart showing the reverberation processing method of the present invention. First, a high-quality filtering module 225 and a low-quality filtering module 250 are arranged (step 201), and then a certain sound quality signal is transmitted. The high quality filtering module 225 is input to generate a high quality reverberation within a certain finite time (step 203). At the same time, the sound quality signal is input to the low quality filtering module 250. Reverberation is generated (step 205) and then the reverberation generated by the low quality filtering module 250 is delayed (step 207), thereby providing the user with the high quality filtering module 225 within a first time period. Only the reverberation is heard. Finally, the high-quality reverberation and the low-quality remnant output from the high-quality filtering module 225 and the low-quality filtering module 250 are output. Immediately after or after the finite time, the user can listen to the low-quality reverberation, and the provision time of the low-quality reverberation is already after the user's identifiable time has elapsed. As a result, the user can feel reverberation, but since it is impossible to identify the reverberation, it is no longer possible to deteriorate the listening quality. At this time, since the reverberation is low-quality reverberation, the amount of reverberation processing is reduced, so that the processing speed is increased and the reverberation processability can be improved.

  As described above, the reverberation processing method and the reverberation processing device according to the present invention are based on the principle that the listening discrimination ability in a certain space of human beings is effective within a finite time, and the high-quality filtering module The excellent points of both the low-quality filtering module and the low-quality filtering module can be combined to provide a high-quality sound effect, and the amount of computation can be greatly reduced, so that reverberation processing can be favorably performed. In addition, the design of the electric circuit can be greatly simplified as compared with the prior art, and the cost of equipment and processing can be greatly reduced.

  Also, what has been specifically described above is merely a working example of the present invention, and all modifications having homologous effects carried out based on the gist of the specification and claims of the present invention. Needless to say, changes and partial diversions should be delivered within the scope of the present invention.

It is a simple block diagram which shows a finite impulse response filter of a prior art. It is a simple block diagram which shows an infinite impulse response filter of a prior art. It is a simple block diagram which shows the reverberation processing apparatus of this invention. It is a curve figure which shows the relationship between the signal of this invention, and time. It is a flowchart which shows the reverberation processing method of this invention.

Explanation of symbols

200 common input 225 high quality filtering module 250 low quality filtering module 275 delay unit 280 adder K delay time T finite time (TK) stacking time

Claims (14)

Step one of placing a high quality filtering module and a low quality filtering module,
Generating a high quality reverberation in a finite time by inputting a sound quality signal into the high quality filtering module;
Generating the infinitely long low quality reverberation by further inputting the sound quality signal into the low quality filtering module; and
Delaying the reverberation produced by the low quality filtering module; and
A reverberation processing method comprising: step 5 of combining the high-quality reverberation output from the high-quality filtering module and the low-quality filtering module with the low-quality reverberation.   2. The reverberation processing method according to claim 1, wherein the step of arranging the high-quality filtering module comprises arranging a finite impulse response filter.   The reverberation processing method according to claim 1, wherein the step of arranging the low-quality filtering module comprises arranging an infinite impulse response filter.   The reverberation processing method according to claim 1, wherein step 4 of delaying reverberation generated by the low-quality filtering module is executed by arranging a delay unit.   In the step 5 of synthesizing the high quality reverberation and the low quality reverberation output from the high quality filtering module and the low quality filtering module, the high quality reverberation and the low quality reverberation are stacked. The reverberation processing method according to claim 1, further comprising:   Step 5 of synthesizing the high-quality reverberation and the low-quality reverberation output from the high-quality filtering module and the low-quality filtering module is characterized in that the synthesis is performed by an adder. The reverberation processing method according to claim 1.   In step 2 of generating a high-quality reverberation within a finite time by inputting the sound quality signal to the high-quality filtering module, the finite time indicates the previous 50 ms. The reverberation processing method according to claim 1. In a reverberation processing device that inputs a sound quality signal by installing a common input terminal,
A high quality filtering module connected to the common input for receiving the sound quality signal and generating high quality reverberation in a finite time;
A low quality filtering module connected to the common input for receiving the sound quality signal and generating infinitely long low quality reverberation;
A reverberation processing apparatus, comprising: a delay unit connected to the low-quality filtering module for delaying the low-quality reverberation generated by the low-quality filtering module for a predetermined time.   The reverberation processing apparatus according to claim 8, wherein the finite time indicates the preceding 50 ms.   9. The reverberation processing apparatus according to claim 8, wherein a finite impulse response filter is used as the high-quality filtering module.   9. The reverberation processing apparatus according to claim 8, wherein an infinite impulse response filter is used as the low-quality filtering module.   9. The reverberation processing apparatus according to claim 8, wherein 25 to 45 ms is adopted as the delay time.   9. The reverberation processing apparatus according to claim 8, wherein the high-quality reverberation and the low-quality reverberation output from the high-quality filtering module and the low-quality filtering module are generated by being stacked. .   9. The reverberation processing apparatus according to claim 8, wherein the high-quality reverberation and the low-quality reverberation are generated by synthesizing with an adder.

Images