Search Gear

Post Production 101

June 30, 2014

A primer for engineering audio for video

Post-production sound involves the processing and mixing of the audio portion of a video or film project after the picture side of production—including all edits—has been finished. The work may include noise reduction, repair of clipped audio, ambience attenuation or enhancement, equalization, dynamics processing (including de-essing and de-booming), and level and pan adjustments.

In this article, I’ll provide pointers for moving a post-production session along smoothly and getting great results. My tips will be focused on audio post for video productions.

Import and Organize Ask the producer— often the videographer—to confirm the sampling frequency (virtually always 48kHz) and bit depth of the audio files they’ll be giving you, along with the video’s frame rate, so you can set up your DAW accordingly before you import any files.

Confirm with the producer that all the audio tracks you’ll be working on start and end at the exact same time. Each file should preferably include some sort of discrete audio spike occurring at the same SMPTE time and before the actual program start, to allow you to visually confirm all the tracks are synchronized to one another and to picture.

For video documentaries and the like, you’ll typically be given separate audio tracks for dialog, music, and B-roll footage. Request a separate dialog track for each speaker appearing in the film or, if that’s not possible, divide the dialog into discrete tracks yourself. If all the dialog or voiceover (VO) tracks were assembled onto one track, it would force you to automate changes to your signal processing chain every time a new speaker enters the program. Separating the VOs allows you to program static settings for much of your signal processing—a critical factor when deadlines and budgets are tight. Be sure to also request a brief recording of room tone (a few seconds will do) for each speaker in the program. The isolated ambience will make it easier to create a noise profile for each dialog track with your noise-reduction software.

Fig. 1. iZotope RX3 Advanced’s Dialogue Denoiser incorporates an all-new adaptive algorithm for ridding tracks of noise. Auto mode is optimized for treating VO tracks.
Clean Up Before you use equalization or dynamics processing on dialog tracks, apply any necessary noise reduction. iZotope’s new Dialogue Denoiser plug-in (part of the company’s superb RX3 Advanced bundle of noise-reduction software; see Figure 1) hushes broadband hiss and HVAC noise in a heartbeat and is incredibly easy to use. (RX3 Advanced includes other outstanding tools for quelling clicks, crackling noises, electrical hum, clipping distortion, and more.) I like to bounce the cleaned-up dialog to a new track before applying any other signal processing, with one caveat: If the dialog track is noisy and muddy-sounding (often the case with dialog recorded with a shotgun or lapel mic), I’ll temporarily apply some clarifying EQ downstream of Dialogue Denoiser so I can hear the track’s embedded noise and any potential processing artifacts more clearly while I work. Once I’ve got transparent settings for Dialogue Denoiser dialed in, I remove the EQ and bounce the track.

Observe the Specs If you’ll be delivering the final mix for a PSA or other project that’s likely to be shown in movie theaters, be sure to adhere to specifications required by National CineMedia (NCM). NCM presents the advertising (including PSAs) you see in digital theaters owned by AMC Entertainment, Cinemark Holdings, and Regal Entertainment Group, among others.

Fig. 2. iZotope Insight’s loudness meters indicate program level has exceeded a user-defined -24LUFS loudness target by 1.1 dB, as shown by the numerical readout in the top center of the GUI and the top LEDs in the meter immediately to its right lighting red.
NCM requires that any audio you submit be either 24- or 16-bit, and in .wav or .aif format. The media company prefers receiving 48kHz audio in discrete 5.1 and Lt/Rt formats but will accept true stereo files. (Lt/Rt is a stereo downmix from 5.1 surround sound that can be decoded with a Dolby Pro Logic decoder.) The voiceover track should not begin within the first second of program start, and it should end at least one second before the end of the spot. Make sure your peak levels don’t exceed -10dBFS or -24 LUFS (Loudness Units relative to full scale). iZotope Insight (a comprehensive suite of metering tools included with RX3 Advanced) provides loudness meters that are extremely useful when mixing projects bound for theaters or broadcast (see Figure 2). It’s yet another reason why RX3 Advanced should be in every post-production engineer’s tool kit.

Keep up-to-date on the latest news
Get our Free Newsletter Here!

You Might Also Like...

Studio Basics
Show Comments

These are my comments.

Reader Poll

Are you a gear DIY-er?

See results without voting »