Skip to content
English
Go to storyterrace.com
  • There are no suggestions because the search field is empty.

AI Audiobooks

What we offer

StoryTerrace offers AI audiobook production as an add-on service. We don't record the narration in-house - it's produced/generated by our external narrator, Princess De Villa (princessmaedevilla@gmail.com).

The process has four steps:

  1. Determine the rate (based on word count)
  2. Brief Princess and get voice samples approved
  3. Confirm with Princess once the client approves a sample
  4. Receive the finished AI audiobook - only if it's going on Amazon, check it against the ACX requirements (separate section below) before upload

Step 1: Determine the rate

Before contacting Princess, get the manuscript's total word count - the rate is per word.

Currency

Rate per word

GBP

£0.015

USD

$0.02

Multiply the word count by the applicable rate to get the total quote (e.g., 31,455 words × $0.02 = $630).

You'll also need to confirm two things with the client before briefing Princess:

  • Accent/English variant (e.g., US English, UK English)
  • Voice gender (male or female)

Step 2: Brief Princess and request samples

Email Princess (princessmaedevilla@gmail.com) with the word count, accent, voice, rate, and the book document attached. Use this template:

Hi Princess,

I hope you are well. We have a new project we would like to create an audiobook for. The project is [word count] words - [accent, e.g. US English] - [male/female] voice. The book document is attached. Rate: [total rate, e.g. $630]

If you could please send some potential samples that would be great.

Thank you. Kind regards,

Example (filled in):

Hi Princess,

I hope you are well. We have a new project we would like to create an audiobook for. The project is 31,455 words - US English - male voice. The book document is attached. Rate: $630

If you could please send some potential samples that would be great.

Thank you. Kind regards,

Princess will reply with three voice samples to choose from.

Step 3: Client approval and delivery

  1. Send the three samples from Princess to the client and ask them to choose one.
  2. Once the client confirms their choice, let Princess know which sample/voice to proceed with.
  3. Princess records and delivers the full audiobook.
  4. If the AI audiobook is going on Amazon, check the files against the Amazon (ACX) requirements below before uploading. If it's just a standalone audio file, this step isn't needed.
  5. Please note Amazon restricts authors to using its own AI generated third-party software, so you cannot use other AI platforms apart from the approved Amazon ones.
  6. Virtual Voice Program: Amazon offers a KDP Virtual Voice beta program that allows eligible Kindle Direct Publishing (KDP) authors to turn ebooks into audiobooks using Amazon's built-in synthetic voices. 
Amazon (ACX) audio requirements — brief Princess on these

This section only applies when the AI audiobook is going on Amazon/Audible. If the client just wants a standalone audio file (not distributed via Amazon), Princess doesn't need to follow the ACX specs below - she can deliver the audio as normal.

Amazon audiobooks are distributed via ACX (Audible/Amazon/Apple Books). Make sure Princess is briefed on these technical requirements before she starts recording, so files don't need to be re-done:

Requirement

Spec

File format

MP3 only

Bit rate

192 kbps or higher, Constant Bit Rate (CBR), 44.1 kHz

File structure

One separate audio file per chapter/section — never the whole book as a single file

Room tone

0.5–1 second of room tone at the start of each file, 1–5 seconds at the end

Credits

Opening credits and closing credits are each their own separate file

Opening credits must include

Title of the audiobook, subtitle (if any), "Written by [author]", "Narrated by [narrator]"

Closing credits must include

At minimum, "The End"

Retail sample

A separate 1–5 minute sample file, starting with narration (not credits/music), free of explicit content

Section headers

Each file must announce its section (e.g., "Chapter 1") within the recording

Length

Each file under 120 minutes; split longer chapters into parts with a secondary header

Audio levels

RMS between ‑23dB and ‑18dB; peaks no higher than ‑3dB; noise floor no higher than ‑60dB RMS

Narration

Must be human narration — no text-to-speech/AI voices

When briefing Princess, be explicit that each chapter/section, the opening credits, the closing credits, and the retail sample all need to be delivered as separate individual files, named clearly with the chapter/section title.

How to upload the AI audiobook on ACX (Amazon's platform)

Note: the book must already be listed as an ebook on Amazon.com before it can be claimed on ACX.

  1. Log in to the ACX account.
  2. Go to My Projects and select Claim your title.
  3. Find the book, select Claim title (it must exist as an ebook on Amazon).
  4. Select Use my own audio files.
  5. Choose the territories and terms of distribution, then review and accept ACX's terms.
  6. Upload the audiobook chapter by chapter — each chapter/section, the opening credits, closing credits, and the retail sample as individual files, uploaded one at a time.
  7. Upload the audiobook cover (must be 2400 x 2400 pixels).
  8. Submit the complete audiobook for review — ACX's Quality Assurance team reviews it (typically within 10 business days) before it goes live on Audible, Amazon, and Apple Books.

If ACX flags an issue (e.g. missing section header, audio level out of range), the title will need corrected files re-uploaded before it can go live.