This page explains the DSP on the output side. These DSPs mainly perform signal processing for speakers (SP). Many output-side DSPs are used in sound systems* to process multiple SPs to produce a single sound. Although this page describes each DSP feature individually, it will be easier to understand if you understand how the output side of a sound system is designed. Please refer to this page regarding the output side design of the sound system.

*  "Sound system" does not have a general or clear definition,
     but on this page it refers to a system that needs to output a wide frequency range at a high volume.

Xorver (Crossover)

Xorver has two important roles.

Separating signals by frequency band
In sound systems, the roles are often divided into an SP dedicated to low frequencies and an SP dedicated to high frequencies. In this case, the signal going to the low-frequency SP must include only low frequencies, and the signal going to the high-frequency SP must include only high frequencies. The role of Xorver is to isolate this frequency band.​

The GUI shown on the left is used to set the filter. In the graph, the horizontal axis represents frequency, and the vertical axis represents the volume changed by the filter.

For example, in the case of the settings shown on the left, the vertical axis has a value of 0 when the signal exceeds 300Hz, so the filter has no effect and the input signal is output as is. On the other hand, below 300Hz, the value on the vertical axis is negative, so the filter lowers the volume of the input signal before outputting it. As a result, only the low frequency range of the input signal is reduced and output.

As shown in the picture on the left, Xorver consists of a combination of a filter that passes only low frequencies (green: Low pass filter) and a filter that passes only high frequencies (blue: High pass filter).
​
​The diagram below shows the entire system up to the SP on the output side.

Time Alignment
Another important feature of Xorver is time alignment. This is a feature that delays the time passing through Xorver by very small amounts of time. Understanding why such a delay function is necessary requires some knowledge of waves. The reasons for this will be explained as simply as possible below.

​The band where both filters (green LPF, blue HPF) overlap is output from the two SPs.




​
If these two outputs are unluckily out of phase, the volume of this band may drop significantly. (For those who do not know what "phase" means, think of "the phases are not aligned" as meaning "the peak positions of the waveforms are not aligned.")

The time alignment function (delay function) is used to correct this. By setting a delay on one Xorver, you can align the phases of both. (The positions of the peaks can be aligned.) This is as if the position of SP was shifted to the back of only one side.

By the way, it may be divided into 3 bands by Xorver.

Filter (Output Fileter)

The output filter is used to cancel the frequency characteristics of the SP.
SP generally has unique frequency characteristics. To explain it in simple terms, the sound changes just by passing through the SP. Filters generally set characteristics that are opposite to those of SP and attempt to cancel them.

​Graphs are used to set up filters, which is exactly the same concept as in Xorver. In other words, the horizontal axis is the frequency, and the vertical axis shows how much the volume is raised or lowered in dB.

What values to set may be written in the SP's manual, or may be measured by an engineer.

In more detail, the filter also has an inverse characteristic that cancels out the characteristics of the room. There is a room (space) between the SP and the listener, and the room also has its own frequency characteristics.

Gain (Output Gain)

Used to determine the output volume of each SP. However, the volume can also be set using the amplifier (AMP), so it may be up to the system adjuster to decide which one to use.

Delay

This function delays the input signal and outputs it. When SPs are installed in front and behind each other, there will be a slight difference in the time the sound output from both arrives at the listening point. In this case, the sound will sound like a duplicate. You will rarely hear two voices completely separated, but the clarity of the voices will be slightly lost. To prevent this, a delay is used to delay the sound output from the closest SP and to align the arrival times.

In reality, the delay time is often set so that the SP in front (the SP placed near the speaker) arrives a little earlier. This is because humans have the tendency to perceive sound images in SPs where sounds that arrive early are output. In other words, this is done as a device to direct the listener's attention in the direction of the speaker.

The delay is set by the delay time (ms). Assuming that the speed of sound is constant (for example, 340 m/s), the corresponding distance is uniquely determined for each delay time. Therefore, a delay may be set depending on the distance difference between the two SPs.