When your project is at the stage we call ‘Picture lock off’ you are ready to do a sound mix. A sound mix consists of going through the entire sound bed of your project with a fine-toothed comb and ‘sweetening’ all the audio. The audio in your project is important, so make sure to check the precise requirements for your intended delivery format/platform.
When creating a film, the sound is often overlooked, thought of as the poor-cousin of video. The truth of the matter is, is that sound is very important. Sound makes up for at least 50% of what your audience will perceive (arguably more) so it’s very important that you get it right. Recording the best audio, using soundtrack’s appropriately and mixing and sweetening it all in the edit will ensure that you’re silent collection of coloured pixels will be perfectly accompanied by a clear, crispy bed of auditory heaven.
Firstly you might have to do some preliminary work on your audio files before they are ready for tweaking. The first big one is to make ‘stereo pairs’ from your audio. Now, you may have already done this earlier on in your project(and it is recommended to do so early on) but if you haven’t yet due to busy-ness or laziness now is the time. Basically, when you record audio, you’re most likely using multiple sources (for example, the camera’s on board microphone and a lapel microphone). The camera interprets these two microphones as your stereo left and right data. SO when the time comes, you need to pick which one is better, disable the one that’s worse, and make a L and R pair from the best audio.
The second major audio consideration is if you’re using outboard microphone’s to record your sound (as recommended) then you will need some way of syncing the audio and video when it comes time to edit. You will need to do this BEFORE you start editing. An external microphone that is plugged into your camera will automatically be part of the clips you record (as your camera’s onboard software will do all the thinking for you). If your microphone is not attached to your camera in anyway then you’ll end up with a video file (from the camera) that has the on-board sound automatically attached and an audio file from your sound recording device. These two files will need to be synced up so that they will play together. There are lots of programs that can help you do this. A great one is called PluralEyes. Depending on the nature of your problem just put the two files in the same timeline and click ‘Sync’. PluralEyes will use the tiny transient sounds present from the onboard microphone and sync them up (using algorithms and again, magic) to match. This can be a nightmare in a lot of different situations so make sure you understand what you are doing before you start recording audio and video willy-nillily. As a side note, this is what clapper boards are for; they create a loud ‘clap’ sound that you can sync manually with the video. Pretty clever.
So you have your stereo pairs and all your miscellaneous audio has been synced. It’s time to start mixing your project. Below is a quick breakdown of the different types of audio you are likely to have in your project.
Dialogue is arguably the most important audio source in your film (unless your creating some avant-garde silent picture) and as such you’ll want to treat it with the care and dignity it requires. Dialogue (in case you’re not sure) is the words spoken by the people in your film, whether it’s scripted dialogue spoken by an actor or a personal interview. The dialogue you record for your film needs to be clear and loud enough for the audience to hear (see tips for recording below) otherwise it has missed its mark. There are many ways you can tidy up or sweeten your dialogue in post processing (see below) but pure, high quality source material is always the best approach.
Music Covered in greater detail in Soundtrack – Music
FX – covered in Soundtrack FX
Atmos – covered in Sound basics Atmos
With so much audio coming from so many different sources, creating a responsible and clear mix can be a challenge in and of itself. How you balance these different components at any given time will depend greatly on their on screen necessity. Is someone delivering a heartfelt monologue? Keep the sound effects to a minimum, the dialogue loud and clear and some soft violin swells quietly underneath, perhaps rising emphatically as the speech nears its end. Is there a montage sequence of a house being constructed? Keep the music loud and on ‘top’ of the mix with sound effects of chopping, hammering, sawing etc (when seen on screen) mixed in. In a montage, usually music and effects will take priority over dialogue, which is sometimes removed all together to give the inference that people had conversations without actually hearing them. Obviously as your scenes change so to will your film’s audio requirements, so remember to keep in mind what should be heard and when.
It’s hard to compile a one-stop document telling you how loud to make your film but a helpful rule of thumb is to have your absolute peaks occur at -6dB and most dialogue comfortably sitting at -12dB. This is a good ballpark figure but in no ways representative of professional delivery standards: these standards might fly on the web but will not be strict enough for television (both shows and commercials have different specifications to!) or feature films.
There are also some technical tips and tricks you can use to keep your sound bed balanced, explained in further detail below.
Editing doesn’t only involve trimming clips of video to make a film. Film is not strictly a visual medium and as such the sound bed of your project needs to have the same (or greater) care taken to ensure a final polished piece. Some of the video editing tips already mentioned in the Toolkit will be applicable here too and as such, won’t be gone over twice. But there are some things that are unique to the editing of audio, and so, in no particular order, here are a collection of tips and tricks that might help you.
When you first import your audio for working on, chances are it could do with an initial cleanup. Make sure to trim unwanted silence from the beginning and ends of clips to create smooth and seamless transitions between clips. In early passes (during assembly or rough cuts) feel free to leave some silence in as ‘handles’ for safety but the closer you get to a final piece, cut, cut and cut some more!
The audio in your project should be recorded (or created) with the highest level of care to ensure clean and clear, audible content. If, for some unforeseen reason your sound is not up to the standard you would like it to be there are a few things you can do to ‘sweeten’ it. You can’t fix everything, and as previously stated, the best position to be coming from is one with clean audio, but if you absolutely must clean your content up, here are a few useful audio effects you might want to consider using that are pretty standard in most NLE’s and digital audio workstation’s (DAW).
Compression is a tool you can use if your audio is generally too quiet or if it’s dynamic range is too great. Dynamic range refers to the difference between the loudest and quietest parts of your sound. If, for instance, your interviewee spoke quietly for some answers and loudly for others, compression can help to bring the quiet parts up and the loud parts down, overall reducing the dynamic range. Be careful not to crush the dynamic range to within an inch of it’s life, some is natural sounding and OK, but it can help to tame overzealous interview subjects, or to generally boost volume in passages that are lacking.
EQ (short for equalization) separates your sound into its discernable frequencies and let’s you adjust each independent of each other. This can be useful in a multitude of situations. If your audio source has a slight hum al throughout it, some judicious EQing can help to remove it. If it is a high pitched, metallic hiss you can turn down the ‘highs’ or if it’s a deep rumbling wind noise for example you can turn down the ‘lows’. For these purposes you may want to use what’s known as a ‘band cut’ or a ‘band pass’ that allow only frequencies within a certain ‘band’ (or range) to pass through or be cut.
Sometimes when different sounds are fighting for attention, rather than using volume or compression to control them, EQ can be more effective. The human voice sits in a very narrow frequency range and sometimes your soundtrack will bleed into this frequency causing the voice and music to compete. A good way to stop this happening is to turn down some of the offending, clashing frequencies in the music. The human voice (roughly) has a fundamental frequency of 85-180 Hz for a typical adult male and 165-255 Hz for a typical adult female. Limiting these frequencies in your soundtrack (known as ‘ducking’) will help to provide much needed clarity for your dialogue track.
Some DAW’s (and less likely some NLE’s) can intelligently remove unwanted sounds from your recordings. Just provide the software with a neutral ‘footprint’ of the offending sound, like the first few seconds of your recording, and the software will (using science or magic) remove the offending frequencies from the recording. It isn’t a perfect technology but using it sparingly can help to tidy up some otherwise ungainly recordings.
When cutting interviews, try to listen carefully to what is being said, and cut out any extraneous information. If they umm and pause a lot, go through the clip and cut out all those gaps, and join the rest of the conversation back together. You’ll be surprised by how cohesive people can sound once you trim away their nervous quirks. Don’t be too judicious however, as you can end up with rambling Frankenstein robots. EXAMPLE VIDEO? And be careful, the more you slice and dice, the more jump cuts you’ll have to cover with overlay. This is mostly OK in the main body of your film once characters have been introduced, but to have enough time to put a lower third over someone’s interview, they’ll need to be on screen, uninterrupted for a minimum of 6 seconds. So keep that in mind.
Otherwise, prune away. You might even find that you can get really creative without damaging the audio content, by putting the end of someone’s sentence at the beginning for instance. By tidying up these interview clips, your film will flow better and your interview subjects will be much more coherent and appear super intelligent and concise. Everybody wins.
Cross ref. with Sound – atmos
So you made sure you recorded atmos at location- x didn’t you? Because if you didn’t, this is when you’re going to find out. When editing, you may encounter situations where you need to cut to silence, to give a moment of breathing space after a comment or to bridge two locations. If you cut to absolute silence it will be very, very noticeable, so this is where you paste in the atmospheric sound that you (so diligently) collected on location. Keep a sequence with your atmos sample open and then you can readily cut and paste it in as required. Different locations will have different atmospheric sounds so collect and collate all atmos from all locations before editing. EXAMPLE VIDEO?
When out on a shoot you might come across a situation where you just can’t remove yourself from a particular sound. One situation we got ourselves into was filming a presentation out in the bush where there was no electricity. They had their own generator running the lighting, microphones and speakers etc. which made an ungodly noise all through the day and night. We couldn’t turn it off and we couldn’t go anywhere else to shoot.
When an unavoidable situation like this occurs you have one final solution up your sleeve. When on shoot, collect footage of the offending sound in question and when back in your editing suite, use cutaways to ‘explain away’ the noise. If you’re audience watches a clip of an interview with a noisy generator they will be distracted by the noise. If you cutaway to the generator buzzing away and then back to the interview, you’re audience will be more understanding and will begin to forget the noise is even there. It isn’t the most elegant of solutions but at least it’s honest, and it does work!
When cutting a piece of video along with a piece of music (that has a beat or rhythm) a very helpful way to help you decide the rhythm of your cuts is to cut to the beat of the track. Some people without a rhythmic bone in their body might find this difficult at first, but it will help to sync the audio and video that much more and provide emphatic punctuation to actions, comments or transitions. Count along with your music (1, 2, 3, 4, 2, 2, 3, 4, just the way you used to in music class) and use your NLE’s play button (or space bar for ease) to stop the play head exactly on these beats. Then make your cuts using these beats as a guide. Some NLE’s will even offer grid overlays that sync with your track’s tempo to provide a useful visual support structure. Always cutting on the beat can become predictable though so always remember to mix it up with every 3rd or 4th cut, not happening on the beat, just to keep your audience on their toes.
Sometimes you will want to see how your scene feels with music added in but you might not already have your soundtrack complete (or even chosen). In this case, a lot of editors will cut to temporary music. Choose a track that emulates the feel of the scene you are going for and cut your scene to this track. You will get an immediate emotional upgrade from throwing in a hugely commercial track with strong emotional power that can be both a blessing and a curse. On the positive side it will help you see which parts of your scene don’t work in the given emotional spectrum. The bad news is you might get too attached to the piece and not be able to use it later (because of highly prohibitive commercial costs). There are lots of ‘royalty free’ songs though, that you can use for your production that cater to a wide range of emotions at a fraction of the cost of using actual songs. Or you could write and create your own to tap into exactly the emotional vein you are looking for. The main message here is don’t get too attached to expensive fancy things that whilst great, you can never, ever have.