“Sound is 50% of the experience”. That’s the George Lucas quote which so many of us are familiar with, and it’s a phrase which I often think of when carrying out my own audio post production work. He was referring to the experience for cinema goers, and he’s absolutely right. A lot of time, care and attention should be put into the audio for film productions in order to ensure that the sound does full justice to the imagery and helps to take audiences on a journey in the way the director intended. Some would even argue, myself included, that sound can sometimes be even more important than the pictures. This brings me to a question: should every production, large or small, have a proper audio post process, beyond just balancing a few levels? That’s what we’ll consider in this article.
What Do We Mean By Audio Post?
Before we proceed any further, I think it’s a good idea to first define what I mean by audio post production. As its name suggests, audio post production should usually be one of the last stages of a video production, ideally happening once the picture is at least locked so no more changes should occur. Audio post involves a variety of different processes, and the extent to which these are required is dependent on the production itself, but broadly, they include:
Noise reduction, especially on dialogue
ADR
The addition of sound effects and/or foley
EQ, compression and other audio sweetening and processing
Balancing and panning of sounds
Music editing and mixing
That’s by no means a comprehensive list, but it’s the key points. Of course, not all of those processes will apply to all productions. For example, you may not need ADR or foley for a simple documentary, but the other processes might be quite important. Does that mean that there are cases when no audio post is really needed? For example, if you have a video which just consists of somebody talking to camera with a piece of music behind them? Can’t you just adjust the balance of the dialogue and music in the video edit and that will do? Well, there’s a bit more to it than that. Let’s take a look at some important points.
Standardised Levels
A while ago, while attending a training course, the course leader played a series of videos to the class. It was very apparent to me that there was a huge disparity in loudness from one to another. One video in particular started playing about 20dB quieter than the previous, leading the course leader to comment “I always have to turn this one up”. This wasn’t an audio course, so the people running it no doubt had no idea how to correct the issue, but I couldn’t help but wonder how a supposed series of related videos could have such obvious and significant level differences from one to the next. This highlights one simple but important benefit to having at least some sort of audio post process: loudness standards. Even if you’re not producing content for TV or streaming platforms, it’s still a good idea to work to your own loudness standard. I work on some corporate content which is largely for internal use within organisations. For about the last decade, I’ve been mixing this type of content to -18 LUFS. This level works for dialogue-based videos which will mostly be played over mediocre sound systems, or through computer speakers. From the experience of the viewer, there’s uniformity in level from one video to the next, meaning nobody has to constantly adjust the playback level as they watch different videos.
Dynamic Range
The subject of dynamic range ties in closely with the previous point about standardised loudness levels. Generally, the higher the integrated loudness, the less dynamic the mix is likely to be. Dynamics shouldn’t just be considered in relation to the mix as a whole though. Individual elements within the mix need to be considered too. If you’re creating video content which is going to be viewed on mobile devices, it’s generally preferable to rein in the dynamic range of important things like dialogue in order to avoid quiet bits getting lost and being inaudible when heard on small speakers in noisy environments. This makes compression necessary, meaning that even a simple production will require some audio post.
Image: Avid
Equalisation
There are hundreds of ways you can process audio, but another key one, in addition to compression, is EQ. Sometimes, getting the elements of a mix to sit well with each requires careful crafting of their frequency balance. To stick with a dialogue-based example, unwanted resonances, particularly in the lower mid frequencies, can often muddy the sound, adversely affecting the clarity. Using EQ to cut these frequencies, sometimes coupled with a boost in the higher mids can really make a huge difference, improving intelligibility and allowing the dialogue to sit better with both music and sound effects. Also, from one voice to another, selective use of EQ serves to iron out some of the differences in the recordings due to mic placement. Slightly more advanced equalisation includes things like dynamic EQ which can help to get everything gelling in a mix when used carefully.
Noise Reduction
Dialogue recordings often end up containing some sort of unwanted noise. This happens on productions of all levels. Capturing clean dialogue when out on location can be very challenging at times, even when the greatest care is taken to avoid it. Noise from traffic, wind, camera equipment or any number of other sound sources can end up on the recording. The severity of this, along with the type of production, really determines whether or not it needs to be dealt with in post. A period drama with aircraft noise on the dialogue will definitely need to be cleaned up. A news piece where someone is reporting from a busy street might be fine, requiring no noise reduction or perhaps just a subtle adjustment. I get sent lots of suboptimal audio, especially from smaller production companies who perhaps don’t have the experience or knowledge to get the best recordings at source. On occasions, the dialogue intelligibility wouldn’t be good enough if left unprocessed. Fortunately, we have some excellent dialogue noise reduction tools available now. Among these, some of my favourites include Hush Pro, Accentize dxRevive Pro and the updated Dialogue Isolate in iZotope RX 11. Viewers have an expectation that dialogue will sound clean, consistent and intelligible. Any less than this detracts from the viewing experience, making noise reduction a necessity in such cases.
Sound Effects
For the purpose of simplicity in this article, I’m going to cover foley and sound effects in one section. For many types of production, the addition of sound effects is an essential part of creating a compelling viewing experience. We all know that feature films require thousands of sound elements to make up the mix, but even short TV commercials or promo videos can still require several hundred sounds, carefully layered together in order to really bring the content to life. Most video editors will at least add some provisional sound effects during the edit. This helps to get a feel for how the finished version might come across. Some of these effects will be good enough to keep in the final version. Others may be ‘temp’ effects which are essentially placeholders which can be swapped for higher quality, more carefully selected and crafted sounds later on. That is of course if there’s a distinctly separate audio post process. Typically, video editors are not best placed to create highly polished, finished sound. This brings me on to my next point…
Audio Knowledge
Reading this article so far, you might be thinking that some of the processes I’ve described could quite easily be carried out within the video editing software. You’d be right in saying this. Audio post actually starts during the edit, but the extent and complexity of this depends on a few things. Will all of the audio post be carried out by the editor, or will the audio be sent to someone else for audio post? In the case of the latter, maybe the editor doesn’t need to do too much to the sound since it will have its own process anyway. Alternatively, audio savvy video editors may choose to carry out their own audio post. Some excellent tools are available these days in video editing software. A notable mention has to be made here for Blackmagic Design’s DAW, Fairlight, which integrates seamlessly with their DaVinci Resolve video editing software. Fairlight provides a superb range of audio post tools which, in the right hands, are more than good enough to allow most of the processes required in audio post to be carried out to a very high standard.
Creativity
Most of the points I’ve made so far have been technical, but there’s another major factor which is absolutely critical to the success of any production - creativity. Creativity is at the heart of the process of putting together a successful video project, whether it be a straight forward training video, a TV drama or a full length feature film. Quite often, the creative people come up with the ideas and have the vision for how the finished piece should be. On any production, a combination of people with different skillsets is preferable in order to guarantee the absolute best result possible is achieved. We need creative and technical people. This is why teamwork often brings about the best result when working on video projects. There are some people who possess both technical and creative skills, but it’s quite rare that both of those skillsets cover the broad range required to not only make a highly watchable, compelling production, but also one which sounds great, technically and creatively.
Just Because You Can…
If you’ve read this article in full, you may have drawn some inferences from the various points I’ve made so far. In my opinion, audio post is such an important part of any production that it’s not something which can just be skipped. If you’re shooting amazing-looking 4K HDR images with great lighting and rich colour, why would you not want the audio to be up to the same standard? Conversely, even a basic video consisting of just someone speaking to camera with some backing music can be elevated to another level through careful use of audio processing, along with consideration for the overall loudness in order to create something which is suitable for the intended playback medium.
For small to medium productions, and where the video editor has sufficient knowledge of audio, the audio post can often be carried out by them, assuming that they have solid understanding of the various points I covered earlier such as loudness, dynamic range, EQ, noise reduction and sound effects. For anything requiring more in-depth or specialised work, or where the editor does not have the knowledge required to complete the audio to the best possible standard, an audio post professional should be used. Where time and budget permit, this will provide by far the best result, allowing the audio to reach its full potential by choosing the right person for the job. Again, this emphasises my previous point about the importance of teamwork. Use the best, most experienced person available for any given task.
Many modern NLEs offer powerful audio tools, but that doesn’t mean that the person using them knows how to get the best from those tools. It’s likely that their core strength is in shooting or editing, not audio. The maxim applies; “just because you can doesn’t mean you should.” So, to revisit my original question of whether or not all video projects need audio post, I would say that this should absolutely be considered a critical part of the process. Consciously or not, audiences have a high expectation of how something should sound, so let’s keep on creating great audio and continue to strive for the best possible sound on all productions, no matter how big or small.