Creating Blue Dust #5

Nicholas Dunkley
5 min readJan 18, 2023

Blue Dust is a sci-fi audio fiction series that I created and released in 2022. It tells the story of a teenager named Max returning to his old home, a floating mega-city, to find his father and get his old life back. You can listen on any major podcasting platform, or at https://anchor.fm/blue-dust

This blog post will focus on Blue Dust’s post production phase.

Starkly contrasting with the production phase’s fast-paced, open, collaborative feel, the post production phase was solitary and slow.

Now that I had the recordings of each voice actor, I had to sort the best takes. This process is called “comping.” I comped all of the actors using the same software used to record them — Pro Tools. There’s a lot of negative talk about Pro Tools, but for comping, editing and mixing, I’ve found it to be excellent.

After I had comped the best of each take, I split the audio into separate lines and laboriously set out to name each one. Coming up with a naming convention that made sense for my project took some time, and naming each of the 1,000+ lines took nearly a week.

In the end, I used this naming convention:
CHAPTERNUMBER_CHARACTER_Line

A typical line looked like this:
CH22_MAX_He won’t be happy

Using chapter numbers instead of names meant that I could easily sort through them in chronological order. Furthermore, to search any line said by a character in that chapter, I only needed to type “CH22_” or “CH22_MAX_” for all lines said by Max in that chapter, etc.

After I had named each line (known in Pro Tools as a “clip”), I started processing them. First, I removed background hums and other noise using Dialog DeNoise inside Izotope’s RX7. Next, I normalised the dialogue using Quiet Art’s WaveRider to -20 LUFS. Even though I’d originally aimed WaveRider to normalise to 16 LUFS, this created distortion during some dramatic, dynamic moments.

Since I used WaveRider, I didn’t use any compression on the dialogue. I believe part of Blue Dust’s natural feel is that its dialogue hasn’t been tampered with very much. I highly recommend Quiet Art’s WaveRider to anyone who hasn’t heard of it for levelling dialogue.

Next, I bulk-exported each line of dialogue at each stage of processing: raw, denoised and denoised at -20 LUFS, respectively.

Now I was ready to start editing.

I made the mistake of editing EP01 in Ableton’s Live. Although there’s nothing wrong with the DAW, I don’t own it and was instead using a trial license. When the license was close to expiring, I had to export all of the DRY and WET stems to import into a different DAW, which was time-consuming and unnecessary.

I originally used Ableton’s Live because of its fabulous “live” view. I originally had the notion of “performing” Blue Dust, similar to an electronic producer queuing samples, rather than laboriously lining up each clip, SFX or piece of music in the timeline, or “session” view, as would traditionally be done. Although this ended up being infeasible, that was my original reason for using Live.

I edited and mixed EP02, EP03 and EP04 in Reason Studios’ Reason. Although this might not seem like an obvious choice, I had my reasons. Firstly, I have a lot of familiarity with Reason, as well as copious effects. Furthermore, I find using automation in Reason to be simple and intuitive. I can drag an area on the timeline that I’d like to be an automation “zone.” In this zone, the automation could turn on an effect, say a reverb on someone’s voice. When the area, or zone, ends, the effect automatically switches off. I don’t need to fiddle very much with points or toggle things on or off. While editing, I could quickly see on the timeline the automations, by which I mean the effects, that were still running for each character, and make adjustments.

Reason afforded another freedoms. I could stack reverbs and other voice effects into a combinator device, which as far as I know is specific to Reason. In this way, Annie or any other character would become a “package,” complete with with effects that are connected with her. For example, in EP01 Annie speaks to Max at the junk drop, in his bedroom, and later outside the vault. Sometimes she has her cybersuit’s visor on, at other times she doesn’t. Furthermore, sometimes she is far away from Max, so her voice needs to sound more distant. Rather than a standalone dialogue track for Annie that shoots off to a bunch of submixes with reverbs or other effects on them, I could package every reverb and effect concerning Annie inside one combinator device, which I could control in one place using automation. It was a very elegant solution.

When editing each episode, I would start with the dialogue and atmos. When areas would come up that would need SFX or Foley, I’d add them then. I usually wouldn’t proceed until I had the dialogue, atmos, SFX and Foley for that moment in place, since I didn’t like shifting dialogue around to create or fill in gaps later. Instead, I laid up each moment of dialogue, atmos, SFX and Foley as they were needed. In this way, I proceeded slowly, building each chapter little by little. I understand that this is a little like laying the foundations of the house, while also wiring it up and putting in the plumbing at the same time, but it worked for me.

Due to my process, the first pass would take a very long time. On the second pass, I would make minor adjustments, as well as add music. Music added weight to emotional moments. I likened it to the paint of a room, or perhaps the furniture. It changed the feel of a scene, and I often spent hours trying to get the music just right for chapters, or even single moments. With minor exceptions, I’m very proud of the use of music throughout Blue Dust.

When I’d finished, most of my episodes were around -18 LUFS, so I ran them through Auphonic, a fabulous online normalising service, to get their overall level up to the standard for Apple Podcasts, -16 LUFS. And after that, I released them! Of course there were many, many hiccups along the way during the post production phase, but ultimately I think I did a good, or perhaps even a very good, job.

Thank you for reading

In future blog posts, I’ll write about the themes and character connections in Blue Dust.

--

--

Nicholas Dunkley

I write about what fascinates me: creating music and podcasts, ambient music, and learning Japanese.