Game Audio Basics: Pops/Clicks and Fades

by Jack Menhorn
(Part 2)
Pops/Clicks
When working with free or cheap sound effects found on the internet; there is the possibility that some files will contain digital pops or clicks.  If a file is poorly edited; these can often happen at the beginning or end of a file if it does not start or end on 0hz (when looking at the file in some wave editors 0 would be the line running horizontally).  These “pops” can also occur anywhere else in the file and can be identified by a large but short spike in the waveform.

 

These will have to be dealt with as the ear can pick up on them very quickly and easily.  If the issue occurs at the beginning or end of a file fade in or out the offending area will likely cause the waveform to end on 0 and remove the pops. If the pop occurs elsewhere in the file one can usually redraw very short spikes of the waveform in a variety of wave editors. Another option would be to cut the offending area completely and crossfade the remaining sound together.

An example of a “pop” is a little over 1 second into this recording of a freezer: https://dl.dropboxusercontent.com/u/16341299/GANG_UNITY_TUTORIAL/GANG_UNITY_TUTORIAL_01_pops_01.wav

If we zoom in on the waveform we can see a very visible spike:

 

https://dl.dropboxusercontent.com/u/16341299/GANG_UNITY_TUTORIAL/GANG_UNITY_TUTORIAL_01_pops_01.png

This spike is the cause of our “pop” and will have to be removed. Using a pencil tool in Sound Forge Pro one is able to smoothly draw out the spike with no ill effects or artifacts caused:

 

If we listen to the new file it sounds much better without the ”pop”:

If one listens closely, a “pop” can be heard at the beginning of the file as well. In a game context: the machine sound would be used as a looping sound and we will discuss how to make the sound properly loop and deal with the “pop” at the same time in the “Looping” section below.
Fades/Length

 

Rarely are recorded/purchased/delivered sounds the exact length they need to be when implemented into the game. Unwanted silence at the beginning or end of the file are paramount to remove not only for sync issues but also for space concerns. If each audio file in your project has 0.5 seconds of silence at the end of them and you have a modest number of 100 sounds in your game then there is a good 50 seconds of useless audio that at 44.1 kHz .wav format comes to 10+ MB of wasted space!

Checking the “tops and tails” of each file for empty space is half of the length battle. The other aspect is the file ending smoothly. If the file was recorded in a reverberant space, or reverb or other effects with decaying tails were added digitally; the sound might fade out after X number of seconds. By subtly adding a “fast” fade on the file; one can have the file “die out” or end smoothly while saving disk space as well as sonic space. A “fast fade” is also quite useful on percussive sounds like footsteps, punches, knocks, to help remove the “roome tone” or ambient sounds on the recording.

The technique also will work on any sound where you wouldn’t want the reverb to come through in the file so that you may add reverb in yourself using Unity’s built in reverb (which we will get to in the future).

 

Here is an example of a footstep recording that uses a fast fade to remove background noise:

 

The first half is the raw recording and the second half is the same exact recording but with careful edits and fades to remove unwanted content. In the first half of the recording one can hear birds, bugs, flies, movement of my clothes and even the subtle sound of me inhaling very slowly. Most of the offending sounds are not distinguishable in the second half of the file. The frequencies and sounds of bugs, birds, and flies and so on are still there, but covered up by the sound of the footstep and then drastically lowered in volume by the fade.

Here is an image of the two waveforms:

 

Now ideally one would not record subtle sounds like footsteps in the outside in the middle of a summer day, but the time and place were chosen to demonstrate how to remove unwanted sounds from sub-optimal recordings made oneself or found on the internet through the use of editing and fades. Ideally an audio professional would record in an environment that would have minimal unwanted sounds or edit the file themselves to remove negative aspects of the recording while designing the final asset for delivery.

end of part 2

Leave a Reply

You must be logged in to post a comment.