New and Noteworthy
Recent Videos News

Post Reply 
Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Audio Engineering and Getting Started
04-04-2015, 05:17 PM (This post was last modified: 06-03-2017 11:47 PM by Ezekieru.)
Post: #1
Audio Engineering and Getting Started
Audio Engineering and Getting Started

Lately I've ran into several people, friends and strangers alike, who seem to be struggling with successfully mixing the audio of their productions. When I ask what their technique is, it usually boils down to “just playing it by ear,” which means they just mix until it sounds okay to them. While this is something plenty of audio engineers do, this is after they've had years of experience under their belt and know how to properly adjust the properties of the clips to fit into one another.

I feel today, I should contribute a simple, easy way to mix your tracks so everything will have an appropriate level of separation between tracks. I won’t go too deep into the matter myself, given I’m still learning loads about the more intricate stuff, but here’s the technique I use for my own productions.

First, let’s take a look at the tools you’ll be using. Let’s stick to a video-editing program this time around, so pull up your NLE (non-linear editor, e.g. Adobe Premiere, Sony Vegas) and look for these three things.

1. Your timeline. The part of the program where you line up your audio and video clips.

[Image: 4797dbb915.png]

2. Your clips. The audio clips that you need to mix. Dialogue along with BGM or ambiance, and perhaps even some SFX.

[Image: e64349b99b.png]

3. Your audio meter. The most important component, this reads out how loud your audio is.

[Image: 8f14e69dd2.png]

Most people seem to ignore the audio meter entirely and adjust the audio’s levels until it sounds okay to the person mixing. This often leads to the audio not working for all audience members’ speakers or headphones. Perhaps your voice has a heavier bass response and you’re using a pair of headphones that boost those frequencies. Someone with a more flat and balance set of headphones might not be able to hear your performance as well as you can.

This is why it’s important to go off of the meter for mixing, so you can manually adjust the levels to create levels of separation that fit with the shots you've constructed for your video.

Here’s a series of audio elements you should be looking to separate. It will be sorted by loudest at the top, and quietest at the bottom.
  • Loud dialogue (e.g. yelling) and loud stand-alone SFX.
  • Average leveled dialogue and stand-alone SFX.
  • SFX to be mixed into dialogue.
  • Ambiance and room tone.
  • Background music tracks.
Great, so we have our list of typical sounds that are used in any audio or video production. Now, how do we level each of these elements so they can fit nice and snug into the mix? This is where the levels of separation come into play.

Let’s pick the center piece of the video: The dialogue tracks. Now we need to set their average levels to a decibel level that is not too high and not too low.

Decibels, for those of you who are new to audio engineering, are the standard measurement of volume in the industry. 0 decibels, or dB for short, is the absolute maximum point in the audio’s volume before the audio begins to sound distorted. Often it is referred to as “peaking”. So we want to avoid ever getting to that point with any of our audio, no matter how loud you want it, while not making the audio too quiet as well.

[Image: 672e51dcb2.jpg]

In my experience, most dialogue should average to around -12 dB. This gives you about 3 to 9 decibels louder and 3 to 12 quieter to play with for all of your other audio content, allowing you to layer them together and construct a decent mix. Why those numbers?

You will want to have an amount that will not only allow you to separate and mix different elements into one another, but you will want it to be consistent. Best way to do that is divide it up mathematically. And to get the most levels of separation without losing the significance of it would be divisions of three.

So, from 0 dB to -24 dB, that gives you 12 layers to choose from. If we shave off 0 and -3 to stay away from peaking our audio, that’s 10 levels. Most of us will only stick to using 2-5 levels at a time, but those looking to improve and add more into your mix, this is really helpful to have.

From here, let’s take the previous list and give some numbers to the elements.
  • Absolute loudest audio element: -3 dB.
  • Loud dialogue and loud stand-alone SFX: -6 to -9 dB.
  • Average leveled dialogue and stand-alone SFX: -12 dB.
  • SFX mixed to the dialogue: -15 to -18 dB.
  • Ambiance and room tone: -18 dB.
  • Lighter-toned BGM tracks: -18 to -21 dB.
  • Heavy-toned BGM tracks -21 to -24 dB.
See? Now that you have these many different options, once you run into a scene with a lot going on audio-wise, you can level each element to fit into one another without taking away the impact of the dialogue. Given how dialogue is the first and foremost priority in audio and video productions, this is a no-brainer.

Now, the main issue with this would run into keeping your audio waveforms as consistent to these levels as possible. However, to anyone who’s recorded a voice over track can tell you, is that the human vocals tend to be naturally dynamic. We jump between talking loud and soft, and for English speakers, emphasis can cause even a single syllable to be significantly louder than the rest of the recording.

You can address this in different ways. The go-to for a lot of voice actors is a compressor, either a physical one hooked up to their chain of equipment, or a digital one in their audio editing programs (e.g. Audacity, Adobe Audition). Another way is for the actor to practice speaking in a consistent manner, but that will most-likely kill the performance you’re looking for. I find it’s best to pull up the waveform, highlight the trouble spot, and lower the amplitude (volume) of that syllable to a more reasonable level.

One last thing to note is the shot itself and how everything should be playing out. Are the characters close to the camera, or far away? Is the sound effect you are adding close, or not? Should the sound have a sharper impact to the audience, or simply an element to add to the background? All of these questions can help you in constructing the audio of the world around you, and give you a better line of judgment when it comes to selecting the right level of amplitude to the clip.

The best way to adjust the audio on-the-fly would to slice up the audio at the start and end points of the trouble spots, right-click on the trouble spot’s clip, and select the Audio Gain function.

[Image: 1d6deee353.jpg]

From there, you can adjust the total gain of it, or shrink it down by the amount of decibels you want.

[Image: 51c541a182.jpg]

If you want to dedicate a whole track or two of your timeline to BGM or ambiance, and don’t wish to manually adjust every single one, here’s what to do. Find out which tracks you want to do this with, then at the top-left, select the Audio Clip Mixer tab and adjust the slider to the left of the track.

[Image: 1bb18cf522.png]

Hope this helps out all those curious on the subject, and if you have any questions regarding my techniques, please reply and let me know!

[Image: tumblr_o50z6sCEXP1usrgjso2_540.gif]
Visit this user's website Find all posts by this user
Like Post Quote this message in a reply
[+] 7 users Like Ezekieru's post
04-10-2015, 07:47 AM (This post was last modified: 04-10-2015 07:51 AM by SoDA.)
Post: #2
RE: Audio Engineering and Getting Started
Thing I noticed about Sony Vegas that's good to mention.

When you apply effects to an audio track, in Sony Vegas Pro, the effects apply after the volume in that track. To have a final volume to use after the effects, you can either put literally put a volume vst at the end of your effects list on that track, or...

If you put any compression in a vst list before then, you shouldn't fuck with the slider that comes pre vst.

Also, if you're going to have environments that all have a mix of there own, and suddenly all those things you've balance to have sound like legit wind or whatever needs to be turned down you really need this.


If you don't see a blue square in your timeline audio tracks, drag the space between the tracks and the timeline to the right. Blue box is master selector.

Now press B. Those buses can have there own envelopes.

[Image: 30243b853a6bb34fdbb4527808914c24.png]

The bottom right green thing can be clicked to have the volume dial be to the right of it. For the effects to be pre-fader.

My sole aim is for each of my posts to be looked at like...
[Image: tumblr_inline_noroeqirJ31rnvg33_500.gif]
Worry not; the abridged community can be dead to anyone, and rightfully so.
Visit this user's website Find all posts by this user
Like Post Quote this message in a reply
Post Reply 

User(s) browsing this thread: 1 Guest(s)