Dark Souls and Social Interaction - Behind the Scenes Part 2

  1. Part 1 | Conception, Research, Writing
  2. Part 2 | Voice Over and Planning
  3. Part 3 | Video Editing (coming soon)

Well, It's been quite a while since the last behind-the-scenes. Three months, actually—I did that at the end of July.

By that time I had something of a final draft done. Shortly later I went on to try to record a voice over, but I wasn't happy with it. That basically sums up the three months.

It wasn't all so bad though. In that time also finished two other scripts: one for Empathy, and the other more recently on Body Ownership that I think is going to be really cool! So those are pretty much ready to go once this Dark Souls video is done. But let's get to why it took so long.

Repairs and Failed Attempts

First attempt I really rehearsed and did it, but going over the audio I just wasn't feeling it. That, and an audio problem with my Blue Yeti Mic seemed to be getting worse.

From when I first got it, the mic has a faint electrical noise. Faint, but enough to piss me off. I tried different computers, USB cables, etc etc but nothing helped. Up until a few months ago I just dealt with it, removing it with noise reduction.

After that recording session though I figured enough was enough, I'd see if I could get it repaired. So I figured out what the RMA process was and got in contact with Blue. This was thankfully a much more pleasant process than I anticipated. They got back to me quickly, and after some back and forth describing the problem and providing proof, they gave me the go ahead to send them the mic.

So I did, and within a week or so they sent me back a different Working Blue Yeti. I don't think it's brand new, but the sound quality is noticeably better.

Unfortunately this brought me right up to my next roadblock; school started. I'm starting my first year of Graduate school so I didn't have too much time to focus on the video. I did tinker with the script every now and then, trying to improve it.

Then beginning of October I gave the script another go. Even though the mic was fixed, this was somehow much worse a take. So that was frustrating. But two weeks after that I tried again and this went much better. So finally, I had a good voice over take with a good mic.

The following week was spent editing through the audio and finally I have a complete voice over done.

Processing Audio

I do all my audio work in Audition. I've used a bit of Reaper in the past, and sometime I might transition to that, but for now that's just what what I'm most comfortable with. Editing the audio isn't too hard, just tedious.

Example of Voice Over file

My approach is to just go through the lines in order, repeating them a few times to get different takes. What I end up with is one or two big wav files ~40min each.

I've learnt the easiest way to actually work through this is to set up a Multitrack project and import the recorded voice over files. Then I simply scrub through them and slice up the different lines. I decided here what to keep, and what to trash. I also take the opportunity to cut out any gross breathing or mouth sounds where I can. Nobody needs to hear that.

Full Project with effects

Once all of those slices are on a single track, in order, I can apply some effects. You can see in the image above I have some effects loaded into the track with all the line cuts. I'll share images of all my settings so you can see exactly what I did. Do note though that I don't do a full mix of the effects, I lower it to around 65%. There are 4 things I do to process my audio:

  1. Noise Removal
  2. Parametric Equalizer
  3. Dynamics Processor
  4. Multiband Compressor

Noise Removal

I try to isolate sound when recording by doing it in my closet, but sometimes its good to remove a bit of the background noise if there is any. This was especially important with my broken mic, but this time around things weren't so bad.

This is done on the original Voice Over files before working with the Multitrack.

Parametric EQ

For the EQ I use this article I found a year or so ago about warming your voice and improving dictation. It's a pretty exhaustive explanation of all the controls you have and what to aim for, but basically I just up the gain a bit around 200Hz and 3000Hz.

Adobe Audition Parametric EQ Settings

Dynamics Processor

I don't use the dynamics processor too much, in fact I turn the mix for that fairly low, but I got it from this video.

Dynamic Processing settings

Multiband Compressor

Finally, I forget where I learned about the compressor but I really like what I can get it to do for my voice. I think it can add a certain richness, although it's easy to go overboard with this. I think it should only be mixed in about 70%. Even then I'm not confident my mix and tinkering with these settings is the best, it's an ongoing learning process.

Multiband Compressor Settings


So now that I have the audio in a single, organized track, my next step is to build a video around that. I'm not sure if this is the best way, but the way I've been making videos is that I'll just throw this audio in Premiere in a Sequence of the quality video I want (I work in 720p/60fps, square pixel ration... although I'm considering trying to make this next video in 1080p).

Once the audio is in I'll throw in the barebones elements. This includes the blue background I use (which is actually a subtly-moving loop of cells I made in After Effects), the intro, and the end slate stuff.

This is what that timeline looks like:

Bare-Bones Premiere Timeline

I think working directly in the video like this makes it easy to conceptualize the different scenes and what I want to do with them. Recently I've been storyboarding these scenes on paper, figuring out roughly what the visuals will look like and planning what scenes I want to Live-Action (although really Voice over quality is a big reason for that, as rather than re-recording I can just do Live-Action for those parts).

This video I'm going to be doing something a bit different. Recently (just last week at the time of writing!) I started releasing short clips from different videos I've made in the past. I'm thinking that I'll make clips out of the different scenes in this new video before the full thing is finished!

Brain Bites

I'm calling it brain bites; mini tidbits of info on whatever subjects I've covered in my stuff. I've found in the past that I often want to reference my videos, but honestly I'm embarrassed to share them. Maybe the video game stuff doesn't match the context of where I'm sharing, maybe just the overall topic is lame, whatever. But the bits of stuff I do inside the video I really enjoy!

So this way, not only am I releasing stuff more frequently (well, once every two weeks), but I'm revitalizing these cool things I've done in videos that otherwise might not get seen by too many people.

I don't feel too bad that most of these clips will be from old videos, either, because honestly not too many people have seen my stuff :P

So anyways, I plan to make a brain bite or two out of new videos I'm making and release those before the full thing is done. This will also act as a kind of teaser, which I think is kind of cool!

This also will give me a kind of schedule. Whatever other video I make will release on the other weeks that Brain Bites don't come out.

This also also basically gives me things to release for months ahead of time. Right now I have Brain Bites lined up to around March! And then as I make videos within that time, hopefully I can keep up easily with those releases.