[ This article was first published in the May, 2009, issue of
Larry’s Final Cut Pro Newsletter. Click here to subscribe. ]
One of the main reasons I started using Soundtrack Pro for my audio mixes was that it had a filter that Final Cut Pro did not.
Since then, I’ve done hundreds of hours of audio editing in Soundtrack and have discovered a wealth of features that make editing audio in it a breeze. But, for this tutorial, I want to focus on the one feature that got me started with the software: the Limiter filter.
In normal situations, the audio level of most sounds bounces rapidly between loud and soft. This difference, called “Dynamic Range,” is an important characteristic of a sound. A sound with a large dynamic range has a large difference between the loudest portion of the sound and the softest.
The problem is that when audio levels change drastically, they can become hard to understand. For this reason, it is often a good idea to minimize, or limit, the dynamic range (that is, to reduce the amount of difference between the loudest and the softest sounds) of audio you want people to understand — this is especially helpful with speech.
While there is a role for limiting dynamic range in music – for example, to even out the thumps in a bass drum – entire libraries have been filled on the subject of when to use limiting, when to use compression, and when to leave them both alone when recording music.
So, for this technique, we are going to look at how to use the Limiter filter to improve the clarity of human speech for use in video production.
Note: One last thought before we plunge boldly ahead. Audio is known for having as many exceptions to the rule as there are rules. While there are legitimate reasons to avoid using the limiter filter in a dramatic situation, in most cases, applying a limiter filter to all dialog – especially non-dramatic dialog – will go a long way to improving intelligibility.
WHAT DOES THE FILTER DO?
In brief, the Limiter filter decreases the dynamic range of your audio; thus making the audio it is applied to sound louder.
To do this, the Limiter filter “limits” the amount of gain (loudness) applied to a clip based upon how loud the clip is. Softer clips get more gain, louder clips get less. The result of this is that the dynamic range is reduced and the apparent loudness is increased. What makes this filter especially useful is that it does this “dynamically,” that is, on an instant-by-instant basis in the clip, rather than by simply changing the audio level of the entire clip, which is what normalization does.
One of the benefits of using the Limiter filter is that when used properly it guarantees that your audio will be consistently louder, but not distorted. This “distortion avoidance” is a HUGE benefit to using this filter; as opposed to simply dragging the red rubber band higher in the FCP timeline.
SENDING A FILE FROM FINAL CUT PRO TO SOUNDTRACK PRO
Audio mixing is generally done when the video edit is complete. So, when the time comes to move your project from FCP to STP, there’s an easy way to do it — called “Sending.”
What sending does is automatically move all your clips from FCP into STP, without having to resort to an intermediate conversion format, or coalesce all your clips into a single AIF file.
Sending is very cool and, even better, very easy.
To send a sequence, Control-click the name of the sequence in the Browser and navigate to Send to > Soundtrack Pro Multitrack Project. (You could also select the sequence in the Browser and use File > Send to > Soundtrack Pro Multitrack Project.)
You send files to Multitrack Projects that you want to mix. You send files to Audio File Projects that you want to repair. You can only send one file at at time to an Audio File Project. There is no limit to the number of files that can be sent at once to a mix.
This dialog then appears. The default checkboxes are all OK for video, so all you need to do is give the transfer file a name and find a place to store it.
On my system, I have a folder in each of my Final Cut Project folders that I use for storing STP project files – which is what we are creating here. That way, if I ever need to go back and re-mix a project… Ah, let me state this differently… That way, when I always need to go back and re-mix a project, all the files are stored in a place I can easily find them.
WHERE DO YOU APPLY THE FILTER?
Unlike Final Cut which applies effects to individual clips, all effects in Soundtrack Pro are track-based. This means filters are applied to entire tracks, rather than just clips.
While to many Final Cut editors this may seem strange, this approach actually encourages you to think about grouping your audio so that similar audio is on the same track.
For instance, I’ll put all my talking head audio on tracks 1 or 2, that way I know where it is and I only need to use one filter for all the audio on one track. Here’s an article that describes how I organize my audio clips during video editing.
Here are two rules of thumb that can help you decide where to apply a filter:
If you have a limited number of tracks in your project, say less than a dozen, feel free to apply a filter directly to the track containing the audio you want to modify.
If you have a large number of tracks, it is often a better practice to assign tracks to submixes, then apply the filter to the Submix track. This reduces the amount of work the CPU needs to do to process your audio and, generally, provides you with better control over the effect.
Since the operation of the filter is the same whether it is applied to a track or a Submix, I will illustrate how to use the filter by assigning the filter directly to a track.
An aside : I just made a note to myself to write about how to use Submixes in a future issue of the newsletter.
HOW DO YOU APPLY THE FILTER?
My recommendation, at least while you are learning how to use audio filters, is to only apply the Limiter filter to tracks with human speech — narration, voice-over, and talking head tracks. I generally recommend against using it for music or sound effects. Applying this philosophy means that your human speech should always be louder, and easier to understand, than the music or effects that underscore them.
Here’s a spoken audio track, loaded into Soundtrack Pro 2 (though this filter also works the same with Soundtrack Pro).
With the track selected, go to the Effects tab in the left pane, select the Dynamic category on the left, then double-click Limiter from the list on the right. (Double-clicking applies the filter to the track. You can also select the filter and click the Plus key. I’m too impatient for that, I always double-click the name of the filter.)
The Limiter filter interface opens. If you’ve never seen anything like this before, don’t panic! Once you understand how this works, you’ll love it.
Here are three settings I make to every Limiter filter I apply:
The Output Level sets the maximum gain for a clip. In this image, because this track is the only audio track in my sequence, I set the filter so that the loudest ANY part of the clip will ever be is -3 dB.
I use these same settings for all my narration audio.
Now it’s time to adjust the filter to make our audio sound great. Grab the Gain slider and, while playing your clip, drag it up until the Gain Reduction bar starts to bounce between the 1 – 2 dB markings. (I added the red arrow in this screen shot.)
What you are doing is increasing the gain of the clip dynamically. If the level of your audio at a particular point in time is -20 dB, the gain, in this illustration, is raised by 8 dB, so the level of the clip at that point in time is increased to -12 dB (-20 + 8).
If the gain of the clip in the next instant is -15 dB, we add the same 8 dB of gain, so that the level in the next instant is -7 DB (-15 + 8).
If the gain of the clip in the next instant is -10 dB, we add 8 dB of gain EXCEPT that the Output Level says that the maximum level of the clip can’t exceed -3 dB. This means we add 7 dB of gain, and throw the rest away. The amount of gain we throw away is indicated by the Gain Reduction bar at the top.
Our goal is to get the Gain Reduction bar to consistently bounce between 1 – 2 dB, meaning we are raising the level of the clip so that the peaks just touch -3 dB. The actual amount of Gain you use will vary with each sequence, which is why we need to adjust it each time.
You’ll be amazed at how this filter can make your talking heads and narration just soar out over the background, without any risk of distortion.
If you have particular settings that you use over and over, click the Show Presets button in the lower right corner.
To apply a preset, simply select it from the menu that appears.
To create your own preset, adjust the settings just the way you want, then click the plus button in the lower left corner. Give the new preset a name and you can reuse this whenever you want.
If all Soundtrack Pro 2 did was this filter, I’d still use it for every mix. The Limiter filter guarantees that my audio is nice and loud, yet prevents it from distorting.
Note: In order for the Limiter Filter to guarantee no distortion, it must always be at the bottom of any filter stack you create in Soundtrack Pro. In other words, the Limiter filter must always be the last filter applied to a track.
Note 2: Click here for a recent article that describes how to prepare your project audio for a mix, and the best levels to use when mixing.
UPDATE – May 19, 2009
Jim Jordan asks:
I would agree with your use of the limiter filter with SoundTrack Pro. How does the new feature in FCP 6 compare: Modify > Audio > Apply Normalization Gain?
Larry replies: Jim, the normalizing filter raises the level of the ENTIRE clip by the same amount. The Limiter raises the level of the clip dynamically, that is, softer passages are raised more than louder passages.
Both provide protection against distortion.
Dejan Savic writes:
I just wondered if this was not just for the reference level at the start of the tape? I live and work in Switzerland, and the Swiss TV asks for the reference sound to be set at -18 dB; for digital tapes, the mix can exceed this reference sound level by 9 dB, which means that the highest sound level can be -9 dB. (For analog you can add 6 dB, which means that the highest sound level can reach -12 dB). In France, some broadcasters ask for a -20 dB reference sound. Maybe that explains the remark that some programs must be set at -20 dBFS?
Larry replies: Dean, you are correct. The -20 dBFS I mentioned refers to the reference tone, not the audio level of the program.
One Response to Using the Soundtrack Pro Limiter Filter
Thank you very much for this excellent tutorial! That´s what I was searching for soooo long!
A few more questions:
1. I have all my interviews on one audio track in FCP. The audio level of the interviews is quite different. What would be the best workflow? Normalize the audiolevel in FCP and then send them to Soundtrack Pro? As in soundtrack Pro the whole track is affected by the “limiter” it won´t work well if the audio level of the interviews are very different.?!
2. I have read about compressor. As far as I understand it´s quite similair as the limiter – what is the difference? What would you recomend?
Thank you Chris