(sonicArts) convolution in audition

I’ve covered convolution in SoundHack in a previous post. You can also use Adobe Audition for convolution, since it has a convolution-based reverb process. Effects in Audition can be previewed non-destructively from within each effects dialog window, but once you apply the effect the process is destructive (the file is changed – once you save it). The main advantages to using Audition for convolution over SoundHack are that you can preview the effect before applying it, and that you can set a wet/dry mix for the process.

The general procedure is to open a soundfile in Audition, select all or part of the soundfile for processing, choose Convolution Reverb from the Effects | Reverb menu, load an impulse file, set options, preview, then apply the process.

In steps:

  • Open an input soundfile in Audition.
  • Add extra silence to the end of the file to allow for the convolution effect to complete. Convolution will add the length of the impulse file to the end of the input file, minus one sample. You can always add a lot of extra silence and then remove the unneeded part afterwards. Insert the pointer at the end of the file and choose Edit | Insert | Silence.
  • Select the entire input soundfile in the editor (in general, having convolution cut out suddenly would be distracting, but I can’t rule out something interesting).
  • Choose Effects | Reverb | Convolution Reverb.
  • Click on the Load button to the right of impulse menu (in the Convolution Reverb dialog window) and choose your impulse audio file. The impulse audio file should have the sampling sampling rate and bit resolution of the input file.
  • Adjust settings
    • Mix is the Dry/Wet mix of the effect. 100% is full wet, or all processed output
    • Room Size adjust the reverb length based on the impulse file. 100% Room Size will give you the impulse file unaltered. Smaller percentages will reduce the length of the impulse file applied, although not in an obvious linear way.
    • Damping LF/HG adjusts the amount the Low Frequencies and High Frequencies are damped. The percentage settings here appear to be counterintuitive. A setting of 100% is completely NOT damped.
    • Pre-delay adjusts the amount of time it takes to reach full buildup of convolved sound. If you have adjusted your amplitude of your impulse file to provide a buildup, you don’t need to add pre-delay.
    • Width enhances (above 100%) or restricts (below 100%) the stereo field of the convolved output.
    • Gain controls output gain of the dry/wet signal (according to the Mix setting). You can add gain to makeup for amplitude loss due to convolution, but be sure to look at your meters. You should not go into the red.

The setting controls with the biggest effect on the output are the Mix, Gain, and Damping (both LF and HF). You can experiment with these settings to get some variation in your effect. If you have an input sound with sharp attacks, leaving some dry signal in the mix can be very effective at keeping some original articulation.

Note that while documentation for Audition indicates that other reverb effects are also convolution-based, only the Convolution Reverb allows you to load your own impulse. The other reverbs compute/render an impulse for you based on your settings.

Finally, your impulse file cannot be longer than 30 seconds.


Leave a Reply