Sound engineers take really sterile tracks and make them sound natural through audio effects. The most common tools to do this are delay/echo and reverb, and with some know-how you can tweak your own tracks for the better.
All of these effects work to help make audio sound better, deeper, and more natural. They’re also used as aesthetic effects, too, and understanding how they work will help you use them to their utmost potential. You can find them all under the Effects menu of Audacity.
One Threefold Effect
Delay, echo, and reverb are all different aspects of the same process: repetition of a sound over time. An echo is pretty easy to grasp. It’s a repeated iteration of a sound that occurs with a diminished volume and after a brief period of time. Shouting into a canyon or wide space will give you an echo. The sound waves are emanating from your mouth, traveling some distance, bouncing off of a solid surface, and returning back to your ears after a period of time. An echo is a delay. In audio editing terms, though, an echo is considered to be a specific type of delay, one that decays but reproduces the sound otherwise accurately. Delay is a customizable thing, and can alter the sound during each iteration.
Then, we have reverberation. This occurs when echoes build up in an enclosed space and cause a volume swell, which then decays as the sound slowly escapes. A good example of this is when you clap or shout in medium sized room with all of the doors closed. There’s a spike that occurs as the sound builds up, so your initial noise isn’t the loudest that it’ll get. Then, after the buildup, the sound will release slowly. You can think of it like an overlapping echo, where instead of full repetitions with a delay in between, an iteration starts very soon after the sound begins and while it’s still going on.
There are three types of delay patterns in Audacity: regular, bouncing ball, and reverse bouncing ball. The regular delay will have a fixed time period in between each individual interation. A bouncing ball delay will start at the delay time and will occur increasingly quickly, the time dropping between each iteration. A reverse bouncing ball delay will start with quick iterations and a low delay time, then gradually slow down until it reaches the max. This last one is often used in reverse effects, a topic we’ll cover in another article.
The decay amount is the value (in dB) by which each iteration will decrease its volume. Using negative values will increase subsequent iterations’ volume. Pop songs often start with this “building” volume delay in the intro, with a crescendo rising and the artist suddenly breaking the music with a word or phrase.
The delay time (in seconds) is the maximum time between each iteration.
The pitch change is measured in semitones, and this will cause the pitch to shift up (or down, if the value is negative) in each subsequent echo. This is another effect found often in pop.
The last value you can change is the number of echoes to produce. Keep in mind that if you don’t have enough silence at the end of your track, these will be cut short. It’s best to add silence at the end of your track before adding a lot of echoes, and trimming off the extra after you apply the effect.
Unless you want to add a simple and quick echo, use delay. This effect can really help bring out overtones in musical tracks and helps add a limited sense of depth to your audio.
The echo is a shortcut effect to an very long regular delay. You can change the delay time and the decay factor and you’ll instantly get a very long echo past your track. Be sure to add silence at the end of your track to accommodate the echo. You can set the decay factor to 1 (no decay), and you’ll get a loop; the audio will repeat with the delay time, but there will be no drop in volume each time and can go on infinitely.
Reverb’s a little more complicated because it has more to do with acoustics. Reverb won’t add echoes; it’ll start building sound up, allow it to peak, and then release it over a period of time. Reverb can help shape your sound and bring out some overtones, and it really helps make clips sound more natural. This is because you’re essentially re-recording the sound in a virtualized space. A bigger roomsize will make the “tail” of the reverb last longer and make the swell louder.
Reverb time affects the duration of the reverb from swell to release. If you make this value really small, you essentially cut out the tail.
Damping is like the decay amount in the delay effect. It governs by how much the overlapping iterations will be cut. This primarily affects the earlier iterations and the tail, though it can decrease the swell if the value is too high. The lower the value, the more intense the reverb will be.
The input bandwidth changes the ranges of frequencies affected by reverb. Smaller values will make it sound dull and muffled while higher values will affect more frequencies and make it sound more bright or intense.
The dry signal level determines by volume how much of the original sound remains in the reverb. The default value is really low. If you want an instant and hard reverb, raise the value. If you’re mixing the original track with another that has added reverb, keep this value low otherwise you’ll get clipping, which ruins the fidelity of the audio. If this is all the way down and you still get clipping, lower the amplitude of the tracks before you apply the reverb.
The early reflection level changes how the early iterations shape the overall reverb. Lowering this volume will cause fewer early “echoes” and change the integrity of the sound. This is difficult to describe so you’ll have to listen and try for yourself here.
The tail level determines by volume the intensity of the reverb by changing the tail section.
Normally, the early reflection level is 15 dB or so higher than the tail level. If you change it so the early reflection level is lower than the tail level, you create the illusion of distance between the source and the listener. Also, note that reverb depends a lot on stereo audio, so use speakers when tweaking this effect instead of headphones. You can also split stereo tracks and apply different settings for the left and right channels. This creates more distinct channels.
Reverb is great for replication the sound of wide open and large enclosed spaces. You can make tracks sound like they were performed in a deep sewer, a gigantic cathedral, or a simple concert hall, and you’ll affect the tone so it sounds more natural than just adding a really short delay or echo.
Simulating space without having to actually record in it makes it easy to add intricacy and naturalness to your tracks. It works great non-music audio, too. You can use it on podcast audio, for example, with the same effect. Have some experience with these effects? Feel free to share some of your favorite settings and their uses in the comments!