Think that I have a speaker and a microphone. I want to adjust the equalizer automatically depending to the surrounding the system is in with just the speaker and a microphone. How can I do that? Which equipments do I need?

When I say automatically, I meant that I should just click on some pad and then the system should take care the rest.

I am thinking of something like this:

  1. There will be two or more speakers which can also function as microphones.
  2. First, one of the speakers will transmit and other one will listen. Wide range of frequency and amplitude will be transmitted and other one will be listening.
  3. Step 2 will be repeated for vice versa.
  4. Then do some sound-signal processing.
  5. Then equalizer will be adjusted according to gathered data.

I look up on internet for something like this and I could not find something like this. Just a few patent claims.

So I am willing to get into this but I need to be sure of few things.

1- Speakers(also function as microphones) will be still and not allowed to change their positions. So the sound check kinda be done in speakers perspectives. Would this equalizer adjustment still be valuable?

2- If it is valuable, what kind of components do I need?

3- Would it be reasonable to do equalizer-adjustment in real time?


2 Answers 2


To do this manually you need a pink-noise generator and spectrum analyser. Pink-noise is random noise with equal energy per octave - white noise has equal energy per decade. The microphone is positioned at the ideal listening position. The operator then adjusts a graphic equaliser to get the desired frequency response - usually flat.

There will be two or more speakers which can also function as microphones.

This is unlikely to work. Speakers make poor microphones due to poor sensitivity and frequency response, speaker cable doesn't make good mic cable and the speakers aren't in the listeners' position.

2- If it is valuable, what kind of components do I need?

A decent microphone positioned in the audience, a means of transmitting data back to the controller, software to interpret the data and send it to the equaliser device.

3- Would it be reasonable to do equalizer-adjustment in real time?

Maybe, but you would need to compare the feedback with that transmitted and account for time-delay (due to slow speed of sound in the air), phase errors, etc.

  • \$\begingroup\$ Thank you for the answer. But, what about smart phones? Think that I wrote an application to my phone, and made my TV or monitor compatible with the application (via bluetooth or something). Do today's smart phones have enough qualified microphones for listening pink-noise? \$\endgroup\$
    – Alper91
    Jul 23, 2016 at 9:25
  • \$\begingroup\$ @Alper91. No, far from it. \$\endgroup\$
    – winny
    Jul 23, 2016 at 10:39
  • 1
    \$\begingroup\$ They do have the hardware to do rt fft though. If your smartphone supports USB otg you can possiblu buy a nice usb mic. \$\endgroup\$ Jul 23, 2016 at 10:45

I doubt that this can work in a good manner.

Regarding the first point, whether a sound check done from the speakers perspective makes sense. I don’t think so. One example (which is in place in the room I am currently writing from) is if there are obstacles between speakers. This for sure distorts your measurement considerably for all but subwoofers.

Also, speakers are directional. This means that if you position the speakers well, they point to the main hearing position. This means that their microphones would mostly pick up distortions and reflected signals, which would in that form never be heard by a human listener. That would lead to over-compensation.

Regarding the second point: You would need a Digital Signal Processing capable processor. That kind of compensation is best done in software running on suitable hardware.

Regarding the third point: Depends on what you mean by "real time". If you by that mean adapting the equalisation based on changes in the room while a signal is being played, I doubt it. You would have to all but one channel regularily. While it would also to a certain extent and theoretically be possible to make use of disjunct stereo or 5.1 signals, I presume that the information in those signals is not enough to be able to reconstruct a proper equalisation from that.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.