Less Battlestar Galactica, more Me.

August 24, 2016 7:33 PM

I have speech software in my own voice that I want to use to help me narrate an audiobook I'm planning. But I don't have a good understanding of how to adjust the controls and adjust my voice so that it sounds less like a Cylon and more like me.

Is there a tutorial that shows how to make adjustments to these control options so I can fine tune my voice rather than changing things hit and miss, which can and has taken forever?

There is no tutorial with the software. Some attributes are easy enough for me to adjust, like pitch, speaking rate and speaking volume. But other attributes I don't understand still leave me sounding like Microsoft Sam. They include:

F0 Target Cost
Stress Target Cost
Accent Target Cost
Robustness Target Cost, etc.

Any help is appreciated.

posted by CollectiveMind to Technology (4 answers total)

Sorry no thoughts on settings but do have someone else with a decent ear listen, we are all physiologically disposed to have invalid perception of our own voices.
posted by sammyo at 7:53 PM on August 24, 2016 [2 favorites]

What microphone are you using?
posted by zebra at 6:14 AM on August 25, 2016

Synthesized speech in 2016 is good enough to be completely understandable for almost every purpose, but even the best of it never sounds properly human. If you're making an audiobook that you expect people to enjoy listening to, have an actual person record the narration.
posted by flabdablet at 7:43 AM on August 25, 2016 [5 favorites]

I second flabdablet-- audiobooks are acted more than read, that's what makes them enjoyable. If you're planning a technical book, the acting skills might not be as key, but the intonation, emphasis, and pacing will be.
posted by instamatic at 9:59 AM on August 25, 2016 [1 favorite]