Introduction

Editing an audiobook for ACX (Audiobook Creation Exchange) requires more than just cutting silence and removing mistakes. Every file you submit to Audible, Amazon, and iTunes must pass a rigorous quality control (QC) process that checks for consistency, clarity, and technical precision. Knowing exactly what ACX inspectors listen for—and how to prepare your audio accordingly—can mean the difference between a fast approval and a rejection that costs you time and money. This guide covers practical, step-by-step editing practices that help you meet ACX standards reliably, so you can focus on delivering a great listening experience.

Understanding ACX’s Quality Control Requirements

ACX publishes explicit quality control guidelines that every narrator and producer must follow. The QC team listens for problems in three main categories: audio quality, performance consistency, and file format compliance.

Audio Quality Metrics

ACX requires a finished audio file that is free of audible background noise. The noise floor should be low (preferably below -60 dBFS), and the overall recording should not contain clicks, pops, hums, or hisses. The RMS (average) loudness must fall between -18 dB and -23 dB, with peaks not exceeding -3 dBFS. This ensures the audiobook plays at a comfortable, consistent volume across all listening devices.

Performance Consistency

Inspectors check for consistent narration pacing, natural breathing, and correct pronunciation. Any section that sounds rushed, overly paused, or emotionally mismatched to the text can trigger a QC flag. You must also edit out repeated words, mispronunciations, and any mouth clicks or saliva noises that become distracting over a long listen.

File Format and Metadata

Each chapter file must be delivered as an MP3 with a constant bitrate of 192 kbps (or higher), a sample rate of 44.1 kHz, and stereo or mono (mono is strongly recommended for narration). Proper ID3 tags (title, author, narrator, album, track number, genre) must be embedded. Cover art must be exactly 2400×2400 pixels as a JPEG or PNG.

Pre-Recording Preparations That Simplify Editing

Quality editing begins long before you open your DAW. Setting up your recording environment and signal chain correctly saves hours of corrective work later.

Choose a Quiet, Treated Space

Avoid recording in rooms with hard surfaces that cause echo or reverb. Use acoustic panels, foam, or heavy blankets to dampen reflections. Even a well-positioned closet full of clothes can work. The goal is a dry signal that gives you clean material to work with.

Invest in a Proper Microphone and Interface

Use a condenser microphone (like an Audio-Technica AT2020 or Rode NT1) with a pop filter to reduce plosives. Pair it with an audio interface that provides clean preamps. Record at 44.1 kHz sample rate, 16-bit or 24-bit depth, in a lossless format (WAV) during production—you’ll convert to MP3 at the final stage.

Set Consistent Levels Before Recording

Your recording level should peak at around -6 dB to -3 dB to leave headroom. Speak at your normal narration volume and adjust the gain so your quietest parts don’t drop below -20 dB. Consistent input levels make normalization and compression much easier during editing.

The Editing Workflow: Step by Step

Once you have raw audio files—usually one per chapter—follow a systematic editing workflow to ensure every file meets ACX standards.

1. Remove Background Noise

Use a noise reduction tool (like iZotope RX, Audacity’s Noise Gate, or Reaper’s ReaFir) to sample a few seconds of room tone and reduce the noise floor. Be careful not to over-process; aggressive noise reduction can cause artifacts that ACX inspectors will catch. Aim to remove the hum and hiss without making the voice sound robotic or underwater.

2. Edit Out Mistakes and Unwanted Sounds

Listen through each chapter and cut: misread words, stumbles, long pauses (more than two seconds), mouth clicks, breaths that are louder than normal, and any coughs or lip smacks. Use a crossfade or a slight overlap when cutting to avoid audible click artifacts. Do not edit out every breath—natural breathing is necessary for realism. Just remove the ones that are distracting.

3. Comp the Best Takes

If you recorded multiple takes of the same passage, comp (combine) the best sections from each take. Ensure the background noise and tonal quality match across takes; otherwise, use crossfades to blend them smoothly.

4. Apply Mild Equalization (EQ)

A light EQ can remove low-frequency rumble (below 80 Hz) and reduce harshness in the upper mids (around 2–4 kHz). A high-pass filter set to 80–100 Hz is common. Keep EQ adjustments subtle—you want to preserve the natural sound of your voice.

5. Use Compression to Even Out Volume

A small amount of compression (2:1 or 3:1 ratio, with a low threshold) can smooth out variation between loud and quiet parts. Set the attack time to around 10 ms and release to about 100 ms to keep the dynamics natural. Alternatively, use a limiter to catch any peaks above -3 dBFS.

6. Normalize to the Target RMS

ACX requires the overall RMS loudness to be between -18 dB and -23 dB. Measure your edited chapter’s RMS using a loudness meter (many DAWs have one built-in). Normalize or adjust gain so the average falls in that range. The peak level should not exceed -3 dBFS.

Final Output Specifications and Mastering

After you’ve edited the audio, master it for delivery.

Export as Mono MP3

Narrators should use mono for spoken word. Export each chapter as an MP3 file using the following settings:

  • Format: MP3
  • Bitrate: Constant 192 kbps (or higher)
  • Sample Rate: 44.1 kHz
  • Channels: Mono
  • Encode quality: High (CBR)

Double-check that the MP3 encoder doesn’t introduce silence at the beginning or end of the file. Leave no more than 0.5 seconds of silence at the start and 1–2 seconds at the end of each chapter.

Embed ID3 Tags

Use a tag editor (like MP3tag or iTunes) to fill in:

  • Title: “Chapter 01” or the chapter name
  • Artist: Narrator name
  • Album: Audiobook title
  • Track Number: Sequential (1/XX, 2/XX…)
  • Genre: “Audiobook” or “Spoken Word”
  • Year: Publication year

Include the correct cover art file, sized at 2400×2400 pixels.

Validation and Submission Checklist

Before you upload to ACX, run through this final checklist:

  1. Listen to the entire chapter in headphones at a moderate volume. Note any remaining clicks, pops, or unnatural breaths.
  2. Check RMS and peak levels using a loudness analyzer. Ensure average RMS is between -18 and -23 dB and peaks are below -3 dB.
  3. Verify the noise floor is below -60 dB. If you hear a constant hum, reapply noise reduction or re-record the section.
  4. Confirm file format: Is it an MP3 with 192 kbps CBR, 44.1 kHz, mono? Open the file properties to verify.
  5. Inspect ID3 tags for accuracy and completeness.
  6. Check cover art dimensions (2400×2400) and file format (JPEG or PNG).
  7. Run a QC plug-in such as ACX Audio Quality Controller (a free tool) or similar software that scans for common issues.

If you have multiple narrators or a production team, designate one person to do a final QC pass before upload.

Common Pitfalls and How to Avoid Them

Even experienced editors can stumble on certain ACX requirements. Here are the most frequent rejection reasons and corrective actions.

Inconsistent Loudness

If one chapter is louder than another, listeners will reach for the volume control. After editing all chapters, check each file’s RMS and adjust gain so they match within 1–2 dB. Use batch normalization tools if needed.

Breaths That Are Too Loud or Too Frequent

ACX does not require you to remove all breaths, but loud intakes that interrupt the flow must be reduced. Lower the gain of those breaths by 3–6 dB, or use a de-breath plug-in. Keep natural pace intact.

Plosives and Sibilance

Plosives (hard “p” or “b” sounds) create a thump. Fix them by rolling off low frequencies with a high-pass filter or by manually editing the waveform to cut the burst. Sibilance (excessive “s” sounds) can be tamed with a de-esser.

Room Tone Mismatch

If you recorded a retake in a different environment or at a different time, the background noise may differ. Match the noise floor by sampling the room tone from the original recording and pasting it silently under the new take, or re-record the entire chapter to maintain consistency.

Missing or Incorrect ID3 Tags

ACX will reject files that lack metadata. Use a tag editor to double-check before export. If you renamed files, ensure the tags are updated accordingly.

Maintaining a Professional Sound Without Over-Editing

One common mistake is processing the audio too heavily. Over-compression makes the narrator sound flat and lifeless. Too much noise reduction creates a watery or gated effect. ACX reviewers are trained to hear these artifacts. Trust your ears: if the audio sounds natural and clear on decent headphones, it probably passes. When in doubt, compare your edited file to a professional audiobook on Audible to gauge the sound quality.

For additional guidance, ACX provides a production standards page with detailed specs and sample audio files. Review those examples to calibrate your ears to what “professional” sounds like.

Conclusion

Editing an audiobook to pass ACX quality control does not require a million-dollar studio. It does require a methodical approach: start with a clean recording, follow a consistent editing workflow, measure your levels accurately, and validate every file before submission. By mastering these best practices, you not only reduce your risk of rejection but also deliver a polished product that listeners will enjoy without technical distraction. The time you invest upfront in proper editing pays back in faster approval times, better reviews, and a stronger reputation on the Audible platform.