How to design a reasonable reverb in ADI DSP?

Infineon / Mitsubishi / Fuji / Semikron / Eupec / IXYS

How to design a reasonable reverb in ADI DSP?

Posted Date: 2024-01-29

Author: Terry Yuan

Abstract: This article focuses on the detailed description of the requirements, principles and implementation process of reverberation. On the one hand, it can help everyone understand some basic knowledge of the reverberation effect. On the other hand, engineers can refer to these models and use them in their own products to design Algorithms that are more suitable for your own products.

Source of demand for DSP reverb

When sound waves propagate indoors, they will be reflected by obstacles such as walls, ceilings, and floors. Each time they are reflected, some of them will be absorbed by the obstacles. When the sound source stops emitting sound, the sound waves must be reflected and absorbed multiple times in the room before finally disappearing. Therefore, we can feel that when the sound source stops emitting sound, there are still several sound waves mixed for a period of time, that is, the sound continuation phenomenon that still exists after the indoor sound source stops emitting sound. This phenomenon is called reverberation, and this period of time is called reverberation time. .

In order to obtain a high-quality musical effect when performing, reverberation is an extremely important component. As the demand for acoustic-related equipment is getting higher and higher, people have higher and higher requirements for the sound quality in music. The main implementation methods for reverberation include physical simulation, sampled reverberation and artificial reverberation. Physical simulation is difficult to implement in actual scenes due to the huge amount of calculations, and is rarely used. Sampling reverb is simple to implement, but it is not flexible enough and has relatively few types. Artificial reverberation has a small amount of calculation and is simple to implement, so it is widely used in practical applications. Of course, the disadvantage is that it is not as realistic as the first two, but there is no problem in supporting ordinary tuning, mixing, and performance needs. The following will introduce the concept, application and implementation of reverberation in DSP.

The definition and advantages of DSP reverberation

DSP reverb (Digital Signal Processing Reverb) is a technology that uses digital signal processing technology (DSP) to achieve reverb effects. Reverberation refers to the phenomenon of sound waves reflecting, scattering and attenuating in a room or other enclosed space. It can make the sound more spatial, deep and wide. In audio processing and music production, the reverb effect is very important, it can make the sound more natural, rich and three-dimensional. It has the following advantages:

• Flexibility: Reverberation parameters, such as delay time, attenuation rate, room size, etc., can be easily adjusted to suit different application scenarios.
• Real-time processing: Through real-time processing technology, the audio signal is processed in real time to achieve the reverberation effect.
• High quality: Provides high-quality reverberation effect, making the sound more natural and realistic.
• Save resources: Save valuable audio processing resources, such as CPU, memory, etc.

In short, DSP reverberation is widely used in music production, recording, broadcasting, games, movies and other fields. Through DSP reverberation technology, we can create richer, three-dimensional and natural sound effects. Speaking of reverb, one more concept we need to know is echo. Echo is a delayed reflection in one direction, while reverb is multiple delayed reflections in multiple directions. What we can see in the principle of software reverberation is basically divided into the following three types:

• Echo type: An echo system built with multiple echoes. The number of echoes is controlled by itself according to the specific type.
• Impulse response type (IR type): It is often used to collect various models on site and obtain better output effects by convolving with the subsequent sound source.
• Schroeder & Moorer Class: It is a hybrid model structure.

For some of the mainstream reverb types currently on the market, such as room reverb, hall reverb, plate reverb, church reverb, spring reverb, etc., their implementation principles can be implemented using the above three categories. At present, these reverberation types are common among us. In the projects of tuners or mixers, they are mainly used to enhance special effects and increase the atmosphere, space and three-dimensionality of music.

ECHO reverberation system

When talking about echo-type reverberation systems, we have to mention the Comb Filter reverberation device. Simply understood, it is a process in which sounds continuously collide in space and produce echoes. Similarly, on the player side, what we need to play is actually a sound source and a process in which it is appended with countless subsequent echoes, which is called a comb filter reverberator for short. Here we need to establish a mathematical model. The following figure (Figure 1) shows a simple room reverberation model:

Figure 1 Room sound model

As can be seen from the figure, the reflection effect of the house is affected by the size of the room and the intensity of the reflection. If the room is large enough and the sound-absorbing materials are very good, there will be basically no reflections in the room. Otherwise the reflection will be stronger. In room architectural design, the Sabine formula is often used for estimation, and the standard for reverberation intensity is generally RT60. Referring to this physical model, we can derive a series of formulas during the design process of the comb filter, such as:

Suppose the signal spoken by the speaker is x[n]the signal received by the listener at a certain time is y[n]then y[n]What content does it contain?

y[n] should be x[n] + Reflection 1 + Reflection 2 .......

How to express reflection? it should be x[n] delay. We assume that the delay is m, then reflection 1 should be x[n-m] , but we should also consider the attenuation during reflection, which is the reflection effect of the house mentioned above. Assuming that the attenuation is a, reflection 1 should be expressed as x[n -m]*a

So, y[n] = x[n] + a*x[n-m] + a^2*x[n- 2m] + a^3*x[n- 3m]......

Simplify the sum and use the difference or z change to get the difference equation: y[n] = is[n-m] + x[n]

Through the derivation of the above formula, the structure diagram and time domain and frequency domain performance of the model can be obtained as shown in the figure below (Figure 2).

Figure 2 Model block diagram

In the time domain, as a proportional (the feedback attenuation coefficient depends on the attenuation formula designed by itself) attenuation model, it presents a periodic decrease law, as shown in the following figure (Figure 3):

Figure 3 Unit impact response changes with time

In the frequency domain, the system has a periodic response to frequency, and has maximum and minimum values. In this way, we will get a comb-like waveform, as shown in the figure below (Figure 4), so it is also called a comb filter. .

Figure 4 Spectrum and phase performance diagram

From this, we can design a simple algorithm based on such a model. In the DSP chip, its computing power is not very high and the storage space is not very large, but sometimes we need to select a little echo-type reverberation system. When using relatively easy-to-use products, such as some lightweight low-power electronic products, which require a bit of reverberation flanging effect, we can use this method to achieve it. For other products with higher standards and less power consumption sensitivity, we will use the two methods introduced below to achieve better results.

IR type reverberation system

For simulating medium reverberation in real life, just imagine if we are talking face to face in a room, because sound reflections in the room are ubiquitous, during the communication process, there will be the first part of the direct sound. When it enters our ears, its energy is at its highest. Then through various reflections, the energy of the sound is attenuated and slowly enters our ears. This time and energy behave like pulses, so it is described here as one of the impulse response types. ring. So in terms of implementation, how to achieve this reverberation effect that is close to reality?

In the computer field, we often generate IR files based on different reverberation characteristics, or we can also obtain specific spatial reverberation based on recording and other methods. Because there is some reverberation, it is very difficult to implement the algorithm and has certain specific conditions, but when we need this reverberation background, we need to use it.

In terms of implementation, we usually perform convolution operations through specific IR files and original sound sources, and the calculation formulas and methods of convolution are relatively complex. In order to facilitate everyone's understanding, it can be imagined that the input signal and IR are multiplied, so that Achieve the reverberation effect of IR in the input signal.

In terms of DSP implementation, analogous to the characteristic reverb we often see in some host computer software, these IR files will be stored in our Flash in various ways, and may have multiple model1, model2, model3, etc. . Just take out a specific file and perform a convolution operation inside the DSP to output it. This is often seen in specific types of reverberation in some music equipment.

Schroeder & Moorer type reverb system

The ECHO-type reverb mentioned above will have some imperfections after the comb filter is designed. In fact, it can be seen from the amplitude spectrum and phase spectrum that the amplitude spectrum is not flat enough, so under the condition that the resonance peak and transient state are relatively large, the sound performance it brings is very serious, and the phase changes are not constant. Therefore, Schroeder made a lot of improvements to the reverberation technology. In the "Colorless" Artificial Reverberation

#design #reasonable #reverb #ADI #DSP