Audio effects
John Knight tries using Audacity to fix his budget student-film soundtracks the lazy but pro-sounding way.
John Knight tries using Audacity to fix his budget student film soundtracks the lazy but pro-sounding way, and breaks down the tools you need to use.
Worried that your videos sound amateurish? Is the sound dull, or the background noisy? Perhaps you’re a Youtuber who wants to give your videos an edge, or a DIY filmmaker seeking enhanced audio and sound effects. Before sinking time or money into big editing suites, check out what can be done on the easiest editor of them all: Audacity. It’s available on Linux, Windows and macos, with a mature codebase and intuitive interface that has become something of a standard among audio enthusiasts.
If you want to tweak your video’s sound, you’ll need to extract it first and put it back in the video when you’re done, so we’ve included a short guide to this – see the box on the opposite page. We picked Kdenlive for our video editor as it’s by far the most well-known on Linux. For those just wanting some fun stuff to play with, we’ll cover the sillier effects first, then move on to practical effects to help polish your audio.
In its default form, Audacity comes with a limited number of plug-ins under the Effect menu, which we’ll examine first. Scroll down to the bottom of the list and you will probably have some ‘Plug-in’ sub-menus containing a list of extra effects. These were available on most installations we tried, but we can’t guarantee their presence, so we’ll cover these extra effects later.
Most of the effects here follow the same flow of highlighting some audio, opening the Effect menu from the main menu, then choosing your effect. Usually a new window will open with numerous settings and a Preview button, letting you adjust any necessary values before you change your audio. However, not all effects will follow this workflow, and some will require a bit more explanation.
Pitch shifting
Changing pitch enables you to tune audio up or down, so you can make someone sound like either a chipmunk or James Earl Jones. However, unlike the change of sound you get from speeding up or slowing down a record, pitch-shifting does this without changing the tempo or length of the audio.
Enable the effect by selecting your audio and choosing Effect > Change Pitch. The new settings window is broken into two parts: Pitch and Frequency. For the more tonally aware, Pitch is the place to start. Here you alter pitch by starting from a musical note, and then choose the note you want to shift into – whatever target you choose will be displayed in the number of semitones below.
If all of that is gobbledegook to you, just use the Frequency slider. Pull the slider right for something higher, and left for something deeper. A indicator tells you what frequency you’ll shift from and to, as well as the per cent of change taking place.
Silly voices aside, pitch shifting has many practical uses. For instance, it can simulate an older voice, or how someone sounded in their youth. In the musical world it is often used to compensate for changes in tempo and recording speeds, letting a recording be adjusted to fit another performance without changing any notes, or to adjust songs to another tuning.
Wahwah peddles
This effect reproduces the sound of a wah-wah pedal. Famous the world over since the Sixties, this sound is
used mostly by guitarists, where the signal phases in and out to create a funky, trippy sound. Perhaps the most well-known example is the famous guitar riff from
Voodoo Child (Slight Return) by Jimi Hendrix. In a cinematic setting, wah-wah could be used in cheap movies for alien invasions or trippy dream sequences.
To give it a try, select your portion of sound and choose Effect > Wahwah. Playing with the sliders in the Wahwah window, we found the Resonance control had the most immediately tangible effect, while the LFT Frequency slider changed the speed at which the effect phased in and out.
Guitars and B-movie effects aside, wah-wah can add a new dimension to different instruments such as during a bass solo (ask any Metallica fan), or to add a little something to vocal performances. For similar effects, try playing with any phaser or flange plug-ins.
Reverse audio
This effect is pretty obvious: select some audio, click Effect > Reverse, and that portion of audio is reversed. However, be careful where you place this in the waveform. If it’s in the middle of a lump of frequencies there could be a very rough jump in sound, along with a clicking or popping noise. Make sure that you either use this somewhere with some blank space either side, or see the later sections on fades and clicking and popping to clean up the sound.
Reverb
Adds that big resonant sound you get in stadiums, and is often built-in to guitar amps and karaoke machines. If you’ve ever hit the reverb pedal on a piano, you’ll know instantly the difference between a normal piano sound and the massive sound you get with reverb. Music aside, you can use reverb to simulate different environments. For instance, you can take spoken dialogue that sounded otherwise tiny and transform it into something that sounds like you’re in an enormous theatre or aircraft hangar.
To give it a try, select some audio and choose Effect > Reverb from the main menu. In the new Reverb window you can can tweak all sorts of options like Room Size, Pre-delay, Stereo Width and so on. Room Size is fairly self explanatory: a value of 1 may sound like you’re in a bathroom, whereas 100 may sound like a stadium. If you click the Manage button at the bottomleft corner, you’ll find a number of factory presets, including vocal settings, differently sized rooms and a cathedral.
With creativity and experimentation you can create some clever environment sounds. For instance, by turning down the Room Size and Stereo Width, but cranking up the Wet Gain, we were able to reproduce a very convincing sewer pipe environment. Used judiciously, reverb can transform your student DIY movies into something that feels more epic, or perhaps impart some stage presence to a performance otherwise devoid of charisma.
Adding more effects
If you’re keen to play with more effects the Audacity website has a dedicated section for plug-ins, available on the Download page. Audacity has support for LADSPA, LV2, Nyquist and VST plug-ins. Although LADSPA and LV2 plug-ins were originally designed for Linux, many of these effects have been ported to Mac and Windows. However, life is easier for Linux users as your package manager is bound to come up with decent results, and any plug-ins should become available in your extended plug-ins menu.
If the plug-ins aren’t in your menu, they may need to be enabled in the plug-in manager, under Effect > Add/ Remove Plug-ins. Here you can see everything available on the system, whether it be New, Enabled or Disabled.
Fade control
Fades are easy to explain and use. Fades are what happen when a piece of audio starts or finishes with a gradual increase or decrease in volume. Rather than having an abrupt start or finish, it’s a way of gently transitioning in and out of passages, making the experience more pleasant for the listener.
Fades are very easy to do. To fade in, select from the beginning of your audio to the point you want full volume, and choose Effect > Fade In from the main menu. To fade out, select the audio from the point you want it to start decreasing until you want silence, and choose Effect > Fade Out.
Audacity’s approach to fades is admittedly crude compared to something like Ardour, Qtractor or Pro
Tools – those apps have much more advanced
functionality – but at least it’s intuitive. If your fades are too drawn out or abrupt, it may be worth adding more silence to the beginning of the track and moving the fade further up, or applying the fade more than once to adjust the curve at which the volume spikes.
Not only are fades handy for transitions, but they are genuinely useful in cleaning up broken audio, as we will see later in the article. Ultimately, fades are an essential part of any professional’s toolkit as they will make your videos less jarring by avoiding abrupt halts in audio, as well as making smooth transitions and cool exits.
Crossfading
This one isn’t in the main Effects menu, but it is among the extended plug-ins and is probably installed by default. This technique can be tricky to explain, but hopefully our picture will clear things up. Crossfading takes two sound sources and slowly transitions between the two. In a multi-track editor like Audacity this equates to having two tracks with a section of overlapping audio, then smoothly transitioning between the two.
To get started, first you need to have your tracks sitting one above the other, so you can highlight the section of audio at the end of the first track, and the start of the audio in the second. Rather than having one track simply end and then the track below start, there needs to be some overlapping audio so Audacity can fade between the two, so make sure each track has some ‘tail audio’ – a couple of seconds should do. If you don’t know how to move audio, select the Time Shift tool from the main toolbar or press F5.
Now, with the select tool highlight the tail audio of the top track, and while your mouse button remains pressed, drag the pointer down to the second track, highlight the first second or two of new audio, and release the mouse button. To apply the effect, choose Effect > Plugins > Crossfade Tracks.
Audacity will now fade out the top track while simultaneously fading in the bottom track, with each track momentarily running over each other. Crossfading is very common in highlight reels and montages, and allows scene changes without bumps in audio or awkward silences – it is absolutely essential for DJS jumping between songs.
Clicks and pops
If there are sudden loud clicking or popping sounds through your speakers, they’re usually caused by very sudden changes in amplitude – often a result of badly pasted audio. There are two ways to fix this. First, you can try selecting the audio and choosing Effect > Click Removal. However, the lower the threshold you set, the more likely it will have crackly side effects.
For a more certain fix, find the bad part of audio and zoom in closely on the wave. Highlight a portion of the end of the wave, and apply Fade In or Fade Out, depending on which end of the audio you’re fixing.
Noise reduction
Audacity’s noise reduction tool uses a different workflow to most plug-ins and can be initially confusing. First, Audacity needs a ‘noise profile’ in order to analyse the track and know what to cut and what to leave alone. To get started, find and select a blank section of recorded audio that has background noise, such as hiss, ground noise, traffic or whatever is bothering you. Now from the main menu choose Effect > Noise Reduction. In the new window that appears, click the Get Noise Profile button. Audacity will now drop back to the main window.
Now choose all the audio you want to clean, which is likely to be the whole track. Open the Noise Reduction window again (Effect > Noise Reduction), and before touching anything, click Preview to see how the track sounds with the default settings.
If the results aren’t to your liking, start with the ‘Noise reduction (db)’ slider, which turns the effect up or down. It’s tempting to just crank the setting way up, but the more noise reduction you apply, the more a weird ‘swimmy’ sound will appear, along with very dull audio, much like a low-quality web rip. The trick is to strike a balance between removing enough noise for clarity, while still sounding natural.
The Sensitivity slider broadens or narrows the number of frequencies let in. On some test audio near a main road, 0 sensitivity let in all noise, but as we moved the slider right, first the traffic noise disappeared, then the bird noises went, and finally we were left with just the dialogue – albeit very thin and unnatural-sounding.
The last slider, ‘Frequency smoothing (bands)’ is tricky to describe. Turned completely down, it had some very strange artefacts that sounded like sci-fi androids from the ’50s were talking in the background. Turned all the way up, the dialogue sounded more natural, but with a strange phasing effect, almost like wah-wah. However, find the sweet spot and it can really offset the ‘swimmy’ low-fi sound of too much noise reduction. If you’re still not getting anywhere, at the bottom are two choices under Noise: Reduce and Residue. Reduce is the default setting and is what you will have been using so far. However, switch over to Residue and you can hear what Audacity is cutting out. If any of the sound you’re trying to keep is in the Residue audio, back off on the Noise reduction and Sensitivity settings.
Ultimately, if you can strike the right balance between the three sliders, you should be able to improve the clarity of your audio while adding a touch of professionalism.
Bad stereo to mono
There is nothing more amateurish than a video playing sound through only one speaker, but fear not – mono is here to the rescue! You might think mono is a relic of ye olde times, but mono is perfect for certain applications: it’s exactly centred, it comes through clearly on any sound setup, and it’s half the file size.
Mono is an excellent choice for simple spoken monologues (yes, that’s you, vloggers), and any stereo music tracks can run alongside mono tracks in Audacity with no problem. For stereo tracks that have gone wrong, converting to mono will centre the audio, and much of your audience won’t even notice the lack of stereo effects, or even care if they do notice!
There are two methods for mono conversion. For older versions of Audacity, click the track menu (left of the track) and choose ‘Split Stereo to Mono’. You will now have two separate mono tracks, one of which you can just delete. For newer releases of Audacity, from the main menu simply choose Tracks > Mix > Mix Stereo Down to Mono.
Compression and normalisation
Do you have a soundtrack with inconsistent volume? You can regain some consistency with compression and normalisation. But what’s the difference?
Whereas normalisation raises sound levels across the track to something more consistent, compression takes a more aggressive approach to sound dynamics, making quiet parts louder and loud parts quieter. Normalisation has less disdain from sound engineers as it still allows for some realistic dynamics, but compression’s brute force approach is popular with radio stations as it guarantees everything can be heard – very important when you have a weak radio signal.
To normalise your audio, select your whole track by double-clicking the audio and choose Effect > Normalize. The option to normalise your maximum amplitude to -1.0 db is a good choice for dynamic music, but feel free to turn it up to 0 db if you don’t care about dynamics.
To compress your audio, select your whole track and choose Effect > Compressor. We don’t have space to fully cover compression, but the two most important options are ‘Make-up gain for - db after compressing’ and ‘Compress based on Peaks’. The first option cranks the volume right to the edge of the waveform, just before it distorts. ‘Compress based on peaks’ can have weird side-effects depending on whether or not it’s enabled. Use the Preview button to see whether you prefer it on or off.
Be careful with compression – whispers sound like explosions, explosions sound like whispers, and it makes everything sound like a beer advert!
Once you’re finished with your audio production you need to export your work. Under File > Export there is a sub-menu where you can choose to export as MP3, WAV or OGG – or just choose Export Audio if you need a different format.
If you intend to share the audio online use a compressed format like MP3 or OGG, but if you’re about to use the audio in a video editor then use something lossless like WAV audio, as the final product will have compressed audio and video anyway.