How To: Room correction for Stereo Speakers with a DSP

_mg_0561In this article I am going to show you how to leverage a digital signal processor (DSP) to add room correction to your stereo or 2.1 speaker system.

Here’s what you are going to need:

  • a DSP (e.g. the MiniDSP 2×4 with a plugin)
  • a measurement microphone (e.g. the MiniDSP UMIK-1)
  • 2x stereo audio cinch cables (or 3x if you have a subwoofer)
  • REW software installed on a computer

You can use any omnidirectional microphone, e.g. the ones that usually come with AV receivers. In case you don’t have one, I would recommend to buy the UMIK-1, it is factory-calibrated and quite affordable (~75$).

umik

The procedure will consist of the following steps:

  1. measure the frequency response of the speaker system at your listening position (DSP not yet connected)
  2. set a target curve and let REW calculate the equalization / correction filters
  3. use the software that comes with your DSP (in my case the MiniDSP 2-way advanced 21 plugin) to transfer the filter settings to the DSP
  4. add the DSP into the audio signal path between your source and amplifier
  5. verify the result with a final measurement

Let’s look into each step in a bit more detail:

#1: Don’t modify your speaker setup yet, but make sure that all settings on the speaker amplifier are in the default settings (disable equalizers etc). Connect your computer to the amplifier, then plug-in the measurement microphone to the computer and open REW. If you are using the UMIK-1, then REW will ask you on startup if you want to specify a calibration file (you can download that file here).

rew_cal

In REW, click ‘Measure’ and then hit ‘Check Levels’. Pink noise will be played for a few seconds, after that ‘Level OK’ should appear in the window. If not, increase the volume of the speakers and repeat.
Then set the parameters as displayed in the following screenshot:

rew_measure

Now place the microphone at your listening position (at ear level) and start the measurement. Note: For this procedure, it must be quiet in the room and you should not be near the microphone if possible.
After a few seconds, the measured frequency response will be displayed:

rew_measurement

#2: Now go to ‘EQ’ and select your DSP on the right side in the first tab ‘Equalizer’. In the next tab ‘Target Settings’, you can choose the type of correction (full range or just the subwoofer) and the target curve.
Feel free to adjust the parameters as you like. To get e.g. a stronger bass boost, you can for instance increase the ‘LF Rise Slope’.

rew_eq

Once you’re happy with your target curve (blue line), hit ‘Set Target Level’. In the next tab ‘Filter Tasks’, you can adjust the correction range (‘Match Range’). Again, feel free to play with the parameters to find an optimal setting. Click ‘Match Response to Target’ to start the calculation of the correction filters. The result should look similar to the one in the following screenshot:

rew_corrected

You can then click ‘Send Filter Settings to Equalizer’ to export the settings as a txt file.

#3: Now close REW and open up the control software for your DSP. In my case it’s the MiniDSP 2-way advanced 21 plugin. Here you can set several parameters for the signal processing such as input and output gain, crossover (high- and low-pass filter), as well as the parametric equalizers. Open the first ‘Parametric EQ’ block and click ‘Import REW File’ to load the filter settings:

minidsp

Click ‘Copy to Input 2’ to use the same equalizer on both channels (left and right).
In the next step you can set the crossover frequency if you are using a subwoofer. If you don’t use one, make sure that all curves are straight lines (‘Bypass’ all bandpass filters).
In my case, I set the low-pass filter to 85Hz and the high-pass filter to 15Hz for output 3 and 4 (will be used as subwoofer input signal).
For output 1 and 2 (which will be the fed into the amplifier for the left and right channel), I only set a high-pass filter.

minidsp_crossover

If you needed to use more filters, you could specify them in the ‘Parametric EQ – Output x’ for each output separately (e.g. if you only want to set filter for the subwoofer signal). I’m not going to use this and therefore ‘Bypass’ all signals.

minidsp_param

That’s it. Click the green button to connect to the DSP, then select ‘Synchronize’ to download the settings to your DSP.

#4: All you need to do now is insert the DSP into the audio signal path. Therefore connect the audio source with a stereo cinch cable to the input ports of the DSP, then use another cable to connect the amplifier with outputs 1 and 2 and – if you have a subwoofer – yet another cinch cable to connect the subwoofer to the outputs 3 and 4._mg_06462x4_boxed_newNow you should verify that the left and right channel are in the right order (play e.g. this sample video on youtube). In case you hear a mono signal (same output on left and right channel), you might need to select ‘Stereo Mode’ as ‘Sub Output Mode’ in the tab ‘System Settings’.

settings

#5: After that, redo the frequency response measurement that we did at the beginning. This time, the output should look much smoother:

corrected

In my case, the difference before and after was significant: The boomy bass was gone and the sound was much more natural.
Here the measurement before and after in direct comparison:

freq_response_comparison

As you can see, the DSP mainly reduced the high peaks, for certain frequencies by up to 20dB!

Hope this helps to give you a better listening experience and keep your neighbors happy 😉

Leave a Reply

Your email address will not be published. Required fields are marked *