Tag: Behind the Scenes

Happy 23rd Birthday to Barry Anderson 🥳🍇!!

(Even though his birthdate and the year the first novel are taking place are never explicitly stated in the text.  But it says in Chapter 2 that 4th of July is exactly a week after his birthday, so anyone with deduction skillz…)

But in honor of Barry’s birthday, in addition to having a mini party where I eat grapes with my friends, I wanted to post progress on a project I’ve been working on.

One of the weirder but more exciting things I’ve been experimenting with lately is what I’ve started calling “Audio Motion Capture.”

For this version of the audiobook, I record the chapter myself first, so all the acting, pacing, emotions, interruptions, and character inflections are mine. Then I use AI to trace that performance onto different voices.  It’s less “AI reading my book” and more “my voice acting, wearing different costumes.”

Just like visual motion capture makes a performance like Gollum in The Lord of the Rings so full of personality and motivation—because you can still see Andy Serkis behind every expression and movement, while remaining much more immersed than if it were simply a dude in a costume—this method lets every phrase carry the exact emotions and inflections I picture as an author, while immersing you in how the characters would actually sound and how the dialogue should really land.  And I don’t need a giant studio, or an Avatar budget to do it.  It’s a one-woman show, with a mic, Audacity, and a few locally-based AI programs running on my 8gb of VRAM.

This is especially helpful for male voices.  Don’t get me wrong, I’m actually pretty proud of my voicework for (male) Barry, Frank, and Nick, but it’s still pretty obviously a woman recording her best dude voices.  Don’t worry, Nick sounds even more punchable than before.

One of the more exciting, while perhaps controversial, aspects (AI is already controversial, let’s just pile it on here), is that it also means that I can have consistent voices that go with the characters’ cognition.  Along with Faces being consistent with underlying cognition, Justin and I have found that people of the same cognitive types also have body type patterns, and vocal type patterns.  So with this method, each character has a particular stew of voices of real people from their cognition type, blended to the point that it’s not any of those people anymore, but the voices still follow the vocal typing patterns.  I don’t want to give specific examples of the voices I used because the whole point is to create a new voice that’s fitting for the character, not have Barry sound like any one male INFJ(ej) or have Grape sound like any one female INFJ(ej), or have Dania sound like any one ESFJ(ej).

Down below I’ll have the current voices-draft of 1 – Birthday, cut into named chapter sections, the same sections as on the Table of Contents on the chapter itself.  I’ll make a more permanent page for these in the future, as well as, like, pages of all content in timeline order, etc.

I might also include a few other fun clips, even some sneak peeks of later scenes, out of context, etc.

I don’t think I ever shared links to the updated, segmented clips of Chapter 1 in my regular recording voice, and I’ll share those too.  That way if anyone prefers my good ol’ fashioned audio recording, you’re welcome to listen that way; plus you can appreciate my hard work and the fact that I spend hours upon hours getting the takes I want of these things, and editing them down.

But I hope this is exciting to you guys too.  Limitless possibilities here of immersive radioshows, made all by my lonesome!  Hey, if nothing else, I get excited every single time I hear Barry, my Barry, say his lines in his very own voice.

And with it being a story about transformation, liminality, masks, and how make-believe can reveal truth all its own, I think it’s pretty fitting that I can use technological transformation magic to make all these little pieces come to life… while really, it’s all bits of me!  Theseus’s audio recording, I guess.

Much love,
<3 Calise

Cast Rollcall

ChatGPT suggested it’d be fun to do a comparison reel of all the characters saying the same sentence in each of their tones and voices, side-by-side.  I loved that idea and casually recorded all of them on my lunch walk yesterday!  The raw audio is like 5 minutes, uncut (besides a couple pauses in recording because I was walking beside traffic lol), but then the converted voice clip is short, but I’m very pleased with it!

Raw audio:

“Better than Expected” Character Reel

Out of Context Clips

These character audio clips are later in the story, but shouldn’t be too spoilery if you’ve read what’s posted of the story so far, up through Chapter 3 currently.  Although they will have some spoilers if you haven’t read any of the story!  So if that’s the case, go read Chapter 1 first, or listen below in the 1 – Birthday section.

Nick’s Purpose

---

Dania – Run on Sentence of Doom

Here’s the raw takes of a particularly hard paragraph in 2 – Lessons, followed by the character voice version of the final take:

Run-on of Doom pt1:

Run-on of Doom pt2:

Run-on of Doom pt3 😤:

Final take, Dania’s voice:

---

Steven and Grape – “Can’t Live in a Fairy Tale”

---

1 – Birthday

Disclaimer: Currently the character voice version has AI Calise for the narration on most of the tracks, which sounds pretty darn good, imo.  And yet, it does lose nuances about my vocal performance, so I intend to go back through and do versions with my original narration spliced with the character voices.  I wanted to do it today, but I bit off way too much, lol.  I am eternally extra, sigh.

Make-a-Wish Morning

Original:

Character Voices:

9:42am

Original:

Character Voices:

Glitter Baby

Original:

Character Voices:

Why Do You Want to Tell Everyone?!

Original:

Character Voices:

Flip Book

Original:

Character Voices:

Fairy Name

Original:

Character Voices:

Frank’s Reaction

Original:

Character Voices:

No “Off-Switch”

Original:

Character Voices:

Stairs

Original:

Character Voices:

Purple Stranger in the Mirror

Original:

Uh, so this section has no dialogue 😆so it doesn’t really need a new version.  I found one word that was funny to have in a voice, so I guess I’ll still upload the edit below.

Character Voice?:

Conductor’s Baton/Back for Now

Original:

Character Voices:

Clip from 2 – Lessons

Lastly, I love how this section of Ch2 turned out, Driving Freedom, both my original audio, and with the character voices:

Driving Freedom

Original:

Character Voices:

---

Huzzah!  Leave me a comment to let me know whatcha think!  Have a nice day!  I’m going to go eat grapes and hot dogs now, in honor of my sweet, crazy, sometimes-boy.