Reading Aloud 15: Choices & Compromises while recording Rude Mechanicals

When Bill Schafer at Subterranean Press asked me to read Kage Baker’s Rude Mechanicals, I was delighted, because I love the Company stories. I was delighted until I started reading the manuscript and realized that the point of view character was male. I skimmed forward, just looking at dialogue. Most of the characters were male.

I don’t mind doing some cross-gender voicing, but generally avoid it with the POV character, because I think it is confusing for most listeners. I agonized and then emailed Bill and told him that I thought he should hire a male voice artist, because that would serve the story better. He disagreed, and since I really wanted to read it, not much arm twisting was needed.

As I read the entire manuscript, instead of skimming, I realized why he wanted a female narrator. Ms. Baker uses direct address to the audience in a couple of places, so while the narrator stays with Lewis, it is clearly a separate narrative voice as opposed to an extension of Lewis. Know what I mean? So choice number one, was to have a female narrator.

This left me the freedom to pitch the narrator up, above my natural speaking voice. I also chose to make it very feminine to contrast with all the boys running around.

For Lewis and Joseph’s voices, I ran into some trouble. Joseph has more speaking time in some scenes than the narrator. Now, in the stories, Joseph is described as a bass baritone. Clearly, I wasn’t going to achieve that naturally, so we had to look at compromises.

Lewis was the less vocally dynamic of the two, so placing him at the bottom end of my range was easy; I didn’t need a lot of room to hit his emotional levels since he’s a steadier character. Joseph, our bass, on the other hand is very volatile and he talks a lot. I found that I could either nail the character or the pitch, but not both. When I pitched him down, he wound up sounding angry and dangerous, because of the audible effort involved in keeping my voice low. It doesn’t sound strained as if I were going to hurt myself, but the strain is nevertheless present as a tension that was inappropriate to the character. Most troubling, he wasn’t funny. Joseph is very funny in Ms. Baker’s story.

So after recording a test chapter with a lower Joseph, we decided to go back to the higher one because, aside from the pitch, that voicing was truer to the character.

It is true that we could have pitch-shifted my voice to get it to the right range. The software to do that now is good enough that if the voice is heard out of context, it’ll pass as real. However, in the context of the other voices I was generating, the pitch shift was obvious. Why? Because there’s this thing your brain does with a familiar voice, called psycho-acoustics, which basically waves a flag saying “Wrong! Something is wrong!” It’s a complex series of things that involve overtones, positioning, and other technical things that you have no idea that you are processing, you just know that it’s wrong.

To demonstrate, I have three clips for you.

The final Joseph choice.

Me, lowering Joseph naturally.

Joseph, pitch-shifted down 10% from the first clip.

See, even down 10% he doesn’t sound like a bass, but he sounds weird. The weirdness is even more apparent if it’s in the context of an entire chapter of natural voices.

The pitch-shifted Joseph, in context.

With all the other voices that are obviously generated by me, pitch-shifted Joseph sounds like someone else and is jarring. Given those choices, we went with the first voicing, feeling that the characterization was stronger there.

At some point, in a reading, you’ll probably have to face a similar choice and I think that you should go for the voice which will give you the most emotional range and be truest to the personality, even if you have to sacrifice some of the physicality.

Did you know you can support Mary Robinette on Patreon?