Recently, a sound design forum that I belong to debated on what the audio levels should be in a film.
I, of course, chimed in. I was surprised that there were so many different opinions. The group is a good cross section of the sound design community being made up of amateur, prosumer and professional participants. However, despite this eclectic group, there was no definitive answer. There were some guidelines and a general understanding, but still no definitive answer. So, how do you go about mixing sound to picture? I'm glad you asked!
Mixing is an art. Anyone who truly understands this side of the business knows and respects this statement. A top-shelf Hollywood mixer can make thousands of dollars a day when working on a film. Why? They simply can do what no one else can do. They ‘hear' a good mix with trained ears that most people don't have - much in the same way that a cinematographer can ‘look' through a lens and see what most people can't see. But, don't be dismayed. There are some simple tricks and guidelines to follow that will make your production sound as if you went to the pros for help.
There are two undisputable rules to follow when mixing sound. When these rules are followed, you are guaranteed to have a mix that will seem flawless to your audience and keep them focused on the story rather than the technical aspects of the production. If your mix strays from these rules, stop what you are doing and start over again!
Rule #1 - Intelligibility
The first and most important rule of mixing for picture is Intelligibility. Unless it's crowd walla or deliberately filtered for effect, every line of dialogue should be heard and understood. If not, then why have a script?
When I mix for picture I always start with the dialogue tracks. I use these tracks as the standard and mix everything around them - not against them. How loud should the dialogue be? Some mixers would answer in the voice of an artist - "How blue should the sky be in a painting?" Others will be more specific in their line of thought. I am one of those ‘others.' Here is my opinion. Typically, a film's dialogue lives around -24db to -18db. Television and video projects are even louder - as much as -12db. Radio levels are through the roof. The reason for the differences in levels is because it all depends on the environment that the audience will be in when they are listening to the program material.
Quiet theatres with padded walls covered in sound-absorbing fabrics help control the environment. Therefore, the film's sound mix has a wide range of dynamics (the differences between quiet and loud sections of the soundtrack). This is very different from the television listening environment. Typically, television is watched in a house with a host of sounds to contend with such as refrigerators, air conditioning units, aquarium filters, traffic outside, lawn mowers, screaming kids, etc. So, there is a bit of a limit that you can have in the dynamics of the production. Have you ever noticed that a movie playing on TV seems really low in volume? There is nothing wrong with the film's mix, but the environment in which the film is being heard. The environment that radio is listened to in is primarily a car. This means that in order for the DJ to be heard, they have to crank their levels in order to cut through the car's engine, passing traffic, etc.
In order to begin to mix the sound for your project you must answer one simple question: What is the environment that the audience will be in when they hear the project? That will be the key factor in determining the base level for the dialogue. Remember, dialogue is king and should never be sacrificed. Music, Foley, sound effects and all other sonic elements are expendable, but the dialogue should always be heard and understood.
When possible, listen to the mix in the environment of the intended listener. There is no point in mixing a soundtrack in a sound treated studio with high-end speakers if the audience will be streaming it on-line with their laptop (in which case you can kiss goodbye all those lush tones below 200Hz you worked so hard to achieve). It is for this reason that major feature films are mixed in movie theatres. If you are doing a training video that will be played back over a television on an AV cart in a conference room somewhere, then you should listen to the final mix through a television before signing off on the project.
Rule #2 - Consistency
The soundtrack should flow seamlessly with all of the elements blending as one piece. This rule refers to all of the aspects of the soundtrack. If one instrument in an orchestra is out of tune, the entire work suffers. The same is true of a soundtrack. Here are some quick tips to keep your soundtrack's dialogue consistent:
One of the biggest pet peeves that I have in film soundtracks are ADR punch-ins that don't match the location dialogue. I felt this way as an audience member long before I pursued a career in sound. For me it was so obvious that the two pieces of dialogue did not belong together and were certainly not recorded in the same environment. I still scratch my head at this one.
Here is the typical cause: Location dialogue is recorded on a set that has some reverb with microphone ‘type A' positioned a couple of feet away. Then, the ADR is recorded months later in a sound proof studio with expensive microphone ‘type B' placed nice and close to the actor. When the two pieces of dialogue are cut together the difference in tone and character is like black and white. When this happens, the audience is pulled out of the suspension of disbelief that the filmmakers fought so hard to achieve. I never could figure this out. For me, the solution is quite obvious.
The ADR should be recorded with the same brand and model of microphone, with the position of the microphone being relatively the same distance from the actor that it was on location. Next, the reverb or tone of the room should be matched with a plug-in program like Altiverb. On some projects, I've fought to record ADR in the same location that the production audio was recorded in. Again, the idea is consistency. Cinematographers will take great care when lighting a pick-up scene so that it matches the lighting that took place during production. Sound designers should take the same care in crafting the soundtrack.
There are times when the production audio will vary in volume. Don't be a lazy sound mixer! If the cuts don't match in volume, take the time to process or correct the tracks to keep a consistent volume. If one track is too low in volume, you might notice hiss when the track's gain is increased. Noise reduction plug-ins can make this problem go away relatively unnoticed. If this is not an option, you might need to lower the volume of the louder track and find a balance.
The second buzz killer for a soundtrack is the room tone or background noise between cuts. This can occur when the background sound is constantly changing (e.g. traffic) or is sporadic (e.g. planes). For example, this can happened if you shoot an interview in an office building where the air conditioning unit kicks on intermittently throughout the shoot. When edited, the interview has sections with and without the air conditioning. This is most noticeable if there is no music bed.
What do you do? Use room tone to help blend the dialogue together. Any professional location sound mixer knows to record at least thirty seconds of room tone before they leave a location. To be more specific, they should grab room tone before they leave a set-up, not just the location. This is because one side of a living room (e.g. a set-up near a window that picks up outside traffic) might have a different tone than the one on the other side of the room (e.g. near the kitchen). If room tone wasn't recorded, you may need to get a little crafty.
If I'm in a pinch with no options, I like to find a section of ‘silence' in the dialogue and place that section underneath dry sections by looping and cross-fading copies of the silence. This can achieve a very convincing result. Once you've done this, call the location sound mixer and give them a refresher course on why room tone is so important! Another trick is to add your own room tone via a sound effects library. The chances of you finding the exact same room tone is nearly impossible, but the right ambience track might mask enough of the holes to help sell the effect.
This trick works great when cutting dialogue that was shot outdoors with occasional traffic. Place a similar traffic ambience sound effect underneath the entire piece and voila! For example, let's say you've shot a stand-up with a reporter on a street corner and plan on using the audio to cut B-Roll over. If you edit the audio and cut pieces together when the traffic is not the same, it will be quite obvious to the listener (e.g. a garbage truck is in the middle of passing by behind one line of dialogue and an immediate cut to the next line of dialogue where there is only light traffic in the background).
If all else fails, try a music bed. This is an easier sell if you are the final decision maker for the project. Find a music track that fits the mood and theme of the scene that you are working on. Make sure that the music doesn't steal the scene by taking center stage. Often times, music tracks that are orchestral in nature will have swells in intensity. As the scene progresses, the music builds and becomes louder and more prominent. This can be corrected by cutting and looping a milder section of the music or by using a volume envelope to ‘ride the gain' of the track and keep the music in line with the dialogue.
Following these rules will help guide you toward better mixes in your projects. In the end, mixing is about balance. Aim for a mix that is blended consistently with intelligible dialogue and you'll keep the audience's attention every time.
Next month: Crash Course For Mixing Part 2 - Simple Strategies That Make All The Difference
Ric Viers has worked in the film and television industry for more than ten years. His location sound credits include nearly every major television network, Universal Studios, Dateline, Good Morning America, Disney, and many others. His sound design work has been used in major motion pictures, television shows, radio programs, and video games. In 2007, Viers launched his own label, Blastwave FX, to celebrate the release of his 100th sound effects library. To date, he is considered to be the world’s largest independent provider of sound effects, with more than 150,000 sounds and more than 150 sound effects libraries to his credit. He has produced sound libraries for numerous publishers, including Apple, Blastwave FX, Sony, Sound Ideas, and The Hollywood Edge.
For more information visit www.ricviers.com.