When it comes to audio for video, no one ever seems to notice it unless it’s bad. And if it is bad, it can completely ruin your project. While we always try to record the best production audio we can from the get-go, problems are bound to pop up. In the end, we must assess the audio we’ve got and use our mixing, restoration, and dialogue replacement skills to make it all work. So, here are five post-production tips for better audio in your videos.
Tip 1: Pick the Best Sounding Mic for Each Shot
Once you’ve got your OMF or AAF file imported into your DAW and your session organized, you’ll often find that multiple microphones were used on set. Rather than selecting just one of the mics and using it for the entire film, you should audition each mic on a shot-by-shot basis. For instance, it may be that the shotgun mic sounds the best in the first shot, but in the second shot, a motorcycle drove by that was recorded too loud in that channel. Rather than just “making do” with a loud engine sound in your audio, it pays to check out what the other mics picked up for the shot. You may find that the lavalier that the talent was wearing has a much better vocal-to-motorcycle ratio than the shotgun mic. Now, the sound of the two mics probably won’t match up perfectly, but I’ll go over some things you can try to improve upon, below.
Tip 2: Use EQ to Match Different Mics
EQ is every engineer’s most valuable and versatile tool, and in addition to fixing problems like bothersome low-frequency rumble, or noisy hiss in your audio, it can also be used to make two or more audio clips recorded with different microphones sound more like one another. Continuing with the example above, let’s assume the shotgun mic sounds best in general, and is used throughout most of the scene, and that the lavalier is only brought in when needed on select shots. In this situation, it would make sense to try to match the sound of the lavalier to the the sound of the shotgun. Since the lavalier was much closer to the subject’s mouth than the shotgun, it’s picked up a harsher, more sibilant sound, with more high-end presence. In this case, I would loop the two sections in my DAW, and instantiate an EQ on the lavalier track. I’d then low-pass the lavalier, cutting down on the brightness to better match the sound of the shotgun mic. It’s a good idea to “cycle” through these looped sections a few times, paying attention to the high frequencies in each sound, and dialing-in the low pass until the two sound as close as possible. That’s just a quick and dirty example, and there’s a ton more you can do with EQ to match one sound to another, but it illustrates a good workflow you can use to carry out this task.
Now, there are some tools available tools that you can use to make this task easier. For one, I suggest monitoring each sound you’re working with through a frequency spectrum analyzer to give you a visual representation to help with your EQ adjustments. Most DAWs have a stock spectrum analyzer, or you can use an EQ with a built-in spectrum analyzer, like the FabFilter Pro Q 2. Another cool feature of the Pro Q 2 is an EQ-match function that will attempt to match the two sounds automatically. iZotope Ozone and RX have a similar EQ matching feature. While EQ match is not always 100% accurate, this tool can serve as a great aide to your ear when undertaking this tedious task.
Tip 3: Don’t Be Afraid of ADR
If none of the mics used on set captured usable audio for a particular section, dialogue replacement is really your only option. ADR, or automated dialogue replacement, is the process of replacing dialogue in post production by looping the section that needs to be replaced in a DAW like Adobe Audition or Avid Pro Tools, and having the talent follow along and re-record the lines in question. The goal is to get a cleaner recording with a performance as close to the original as possible. To do this effectively, you’ll want to use the same mic that was used on the shoot and, ideally, you’ll have access to ambient sound from the original location to layer behind your ADR audio, which we’ll get into more in a minute. EQ and reverb can also help to give the impression that the replaced dialogue was recorded on location. Also, it’s inevitable that some or all the ADR you record will not match up with the original audio in terms of timing, and therefore will not match the picture. This, obviously, looks bad (see Kung Fu movies from the ’70s with dubbed English.) Luckily, it can be fixed, and there are several ways to do that. The first way is to manually cut up the ADR audio, usually every few words, and line it up with the original reference recording that you’re replacing by hand. Then, you can manually time stretch any words that are too long, or too short. This can work, but also can be a tedious process, and there are some more automated alternatives to make your life easier. If you’re using Adobe Audition, you’re in luck. Audition has a built-in feature called "Automated Speech Alignment,” which will analyze your ADR, and automatically line it up with the original audio. If you’re using another DAW that doesn’t have this specific feature, like Pro Tools, there’s a great plug-in you can buy called VocALign PRO 4, by Synchro Arts, that does the same thing.
Tip 4: Layer Ambient Sound
As mentioned above, a common issue with production audio recorded in different locations, or with different mics, used in the same scene, is that the ambient sound in the background is often different in each mic. This causes choppiness between cuts, and can make adjacent sections sound different. A way to fix this problem is to layer excess ambient sound from the scene in any gaps that don’t already have it. If no excess ambient sound was recorded, you can use pre-recorded ambience that fits the scene from a sound effects library. While it may be necessary to EQ library ambience to get it to match the production audio, when done properly, this technique can really help sell the scene.
Tip 5: Smooth Out Edits Using Crossfades
When making edits that combine different recordings, as we’ve been discussing, crossfading is an essential technique for ensuring your cuts sound smooth. Whether you’re editing in Adobe Audition, Pro Tools, Logic, or another DAW, you can bet there is a crossfade feature. Any time you’re cutting from one audio clip to another, you should use a crossfade to bring down the volume of the first clip gradually, while simultaneously raising the volume of the second clip. If you don’t do this, you’ll often hear nasty clicks, pops, and jumpiness in the audio that you don’t want.
Thanks for reading, and be sure to ask any questions, or leave any thoughts in the Comments section, below.