ACX Production Platform Settings: A Clear Overview

If you’re recording your first audiobook, the ACX Production Platform settings are the difference between a submission that sails through review and one that bounces back with a rejection email. ACX—the platform that connects narrators and producers with authors for Audible, Amazon, and iTunes distribution—has strict technical requirements, and they’re not negotiable. The good news: they’re also not complicated once you understand what each setting does and why it matters.

Here’s what you need to configure correctly before you record a single word.

The Technical Foundation: Sample Rate, Bit Depth, and File Format

Set these three values in your digital audio workstation (DAW) before you hit record. They determine the quality ceiling of your entire project.

Sample rate: 44.1 kHz. This is the audio CD standard, and it captures the full range of human hearing (up to 22.05 kHz). Some interfaces default to 48 kHz because it’s the video standard—if yours does, change it before recording. Recording at 48 kHz and converting later works, but it’s an extra step that can introduce errors if you forget to dither properly.

Bit depth: 24-bit. This gives you 144 dB of dynamic range, which is generous headroom for voice. The practical benefit: when you apply gain during mastering, you won’t amplify noise floor hiss. A 16-bit recording has less headroom, and quiet passages can develop audible background noise when normalized to meet loudness targets.

File format: WAV or MP3. ACX accepts both, but they serve different purposes. You’ll deliver MP3 files at 192 kbps or higher with a constant bit rate. However, recording and mastering in WAV (lossless) preserves every detail of your performance. Think of it like photographing in RAW before exporting to JPEG—you can’t recover detail that wasn’t captured in the first place.

Why this combination matters: A 44.1 kHz/24-bit WAV file captures your voice with maximum fidelity. When you compress to MP3 for delivery, the encoder has rich source material to work from. The file size difference is significant: a 10-hour audiobook in WAV runs roughly 600 MB, while the same content at 192 kbps MP3 is around 86 MB. That matters for upload time and storage, but it shouldn’t influence your recording format.

Loudness Targets: The -23 dBFS Standard

ACX requires your audio to average -23 dBFS with peaks no higher than -3 dBFS. This is the most common rejection reason for new narrators, so it’s worth understanding what these numbers actually mean.

RMS level (the average): Your recording’s average loudness should sit around -23 dBFS. This isn’t a peak measurement—it’s the average over time. Speech naturally fluctuates between quiet and loud moments, so you’re aiming for the overall average to land at this level.

Peak level (the ceiling): Your loudest moments—shouts, emphasized words, emotional peaks—should never exceed -3 dBFS. This leaves headroom and prevents clipping.

A concrete example: Listen to Project Hail Mary by Andy Weir, narrated by Ray Porter. Notice how quiet moments remain intelligible and intense moments don’t distort. That’s careful loudness matching. Porter’s performance ranges from whispered calculations to panicked exclamations, and the engineering keeps every moment clear without forcing listeners to adjust volume.

How to measure it: Use a loudness meter plugin in your DAW. Most modern DAWs (Reaper, Pro Tools, Audacity) support free options like Youlean Loudness Meter. Set it to measure integrated LUFS, and adjust your gain until the average reads around -23 LUFS (which corresponds to -23 dBFS for this purpose).

The practical workflow: Record with your input gain set so your voice peaks around -6 dBFS during normal speech. Then, during mastering, use a limiter to catch stray peaks above -3 dBFS, and adjust overall gain to hit the -23 dBFS average. If your recording is too quiet, add gain; if it’s too loud, reduce it.

A common branch point: After your first mastering pass, check the loudness meter. If your average reads -18 dBFS, your audio is too hot—reduce gain by 5 dB and re-check. If it reads -28 dBFS, you have room to add gain, but watch your noise floor. If you’re adding more than 6-8 dB of gain, your recording levels were too low, and you may hear hiss in quiet passages. In that case, the better fix is re-recording with higher input gain, not pushing the volume in post.

Noise Floor and Room Tone: The -60 dB Rule

ACX requires a noise floor of -60 dBFS or lower. This is where many home studio recordings fail, especially if you’re recording in a room with ambient hum from computers, HVAC systems, or street noise.

What this means practically: Your room tone—the background noise present when you’re not speaking—must be at least 60 dB quieter than your average speech level. If your speech averages -23 dBFS, your room tone should be at -83 dBFS or lower.

How to check: Record 5-10 seconds of silence in your recording space. Look at the RMS level of that silence in your DAW. If it reads above -60 dBFS, you need to address your room acoustics before recording.

Real-world fixes: If you’re recording in a typical home office, you’ll likely need some acoustic treatment. A portable isolation shield (like the Aston Halo or sE Electronics Reflexion Filter) can help, but the real solution is reducing reflective surfaces and background noise. Moving to a closet full of clothes, hanging moving blankets on walls, and turning off your computer’s fans during recording are all practical first steps.

Why this matters for listeners: A high noise floor creates a constant hiss that’s especially noticeable during quiet passages. When listeners use earbuds on a commute, that hiss competes with the narration and causes listening fatigue. It’s also harder to fix in post-production than to prevent during recording.

A failure-mode warning: If your noise floor reads -55 dBFS, don’t assume noise reduction software will save you. Heavy noise reduction creates a “swimming” or metallic artifact that’s often worse than the original hiss. The better path: identify the noise source. If it’s your computer fan, move the microphone to a different room or record with the computer in another room using a long USB or XLR cable. If it’s room echo, add soft surfaces. Only use noise reduction for minor cleanup, not as a substitute for fixing the room.

The Submission Checker: Your Pre-Upload Safety Net

ACX provides a free tool called the Submission Checker that analyzes your audio files against their requirements. This is the single most useful quality control step you can take before uploading.

What it checks: The Submission Checker evaluates your files for sample rate, bit depth, file format, RMS level, peak level, noise floor, and clipping. It also checks for DC offset and other technical issues.

How to use it: Download the tool from the ACX website, then drag and drop your audio files onto it. It will flag any issues and tell you exactly which requirement you’re failing. This is far better than uploading to ACX and waiting days for a human review that comes back with a rejection.

A common scenario: You’ve recorded and mastered your audiobook, and you’re confident everything sounds good. You run the Submission Checker and discover your noise floor is -58 dBFS—just above the limit. Without the checker, you’d have uploaded and waited for rejection. With it, you know to apply noise reduction or re-record the affected sections before submitting.

The catch: The Submission Checker doesn’t evaluate narration quality, pacing, or performance. It only checks technical specs. A technically perfect file with a lifeless narration will still be rejected by human reviewers.

Verification step: After you fix any flagged issues, re-run the Submission Checker on the corrected files. A clean pass—no warnings, no errors—confirms your files meet every technical requirement before you upload. This is your go/no-go signal. If the checker passes, you’re ready to submit. If it still flags something, don’t upload until it’s resolved.

Export Settings That Save You From Rejection

When you’re ready to export your final files, the specific settings in your DAW’s export dialog matter as much as your recording settings.

For WAV delivery: Export as PCM WAV, 44.1 kHz, 24-bit, mono. Audiobooks are mono—not stereo. If you record in stereo, you’ll need to export as mono or collapse to mono during mastering. The file should be continuous, with no gaps between chapters unless you’re delivering each chapter as a separate file (which ACX allows).

For MP3 delivery: Export at 192 kbps constant bit rate (CBR), 44.1 kHz, mono. Don’t use variable bit rate (VBR)—ACX requires constant bit rate for consistency. Some producers export at 256 kbps for higher quality, which is also acceptable.

Chapter breaks: If you’re delivering a single file with multiple chapters, ACX uses the chapter markers embedded in the file. In Audacity, you can add label tracks at each chapter start. In Reaper, you can use region markers. The key is that your chapter markers must be accurate—listeners navigate audiobooks by chapter, and incorrect markers create a frustrating experience.

A real example: The audiobook The Martian by Andy Weir, narrated by Wil Wheaton, has 32 chapters. Each chapter marker corresponds to a distinct section of the story. If a producer accidentally shifted a marker by 30 seconds, listeners would hear the previous chapter’s ending when they try to jump to the next chapter.

A branch point to watch: After exporting, spot-check your chapter markers by playing the first and last 10 seconds of each chapter. If a marker is off, you’ll hear dialogue from the wrong section. Fix the markers in your DAW and re-export before running the Submission Checker again.

Common Settings Mistakes and How to Avoid Them

Even experienced producers make mistakes with ACX settings. Here are the most common ones and how to prevent them.

Recording at 48 kHz and forgetting to convert. This happens when you use a microphone interface that defaults to 48 kHz (common for video work). The fix: check your DAW’s sample rate before recording, and convert during export if needed.

Normalizing to -3 dBFS peaks. Some producers normalize to the peak limit, which makes the audio too loud on average. Remember: the average should be -23 dBFS, not the peak. Normalize to the average, then use a limiter for peaks.

Submitting stereo files. Even if you record in stereo, ACX requires mono delivery. Stereo files double the file size without adding any benefit for spoken word content.

Ignoring the Submission Checker. This is the most avoidable mistake. Run every file through the checker before uploading. It takes minutes and saves hours of waiting.

A cautionary tale: A narrator recorded an entire 12-hour audiobook at 48 kHz, then exported at 44.1 kHz without properly dithering. The resulting files had audible artifacts—a subtle metallic quality that the ACX reviewer flagged. They had to redo the entire export process, which took a full day. The lesson: check your sample rate before recording, not after.

Final Thoughts on ACX Production Settings

Getting your ACX production settings right is a matter of understanding a few technical specifications and building them into your workflow. The good news: once you’ve configured your DAW correctly, you can save a template and use it for every project.

The requirements aren’t arbitrary—they exist to ensure consistent playback quality across Audible’s vast catalog. When you meet them, your audiobook sounds professional alongside titles from major publishers.

If you’re just starting out, run the Submission Checker on every chapter before uploading. It’s the fastest way to learn what your setup needs. And if you’re considering producing your first audiobook, remember that the technical side is learnable—the performance is what listeners will remember.

Listen to this on Audible — Start your free trial and get two free audiobooks. It’s a low-risk way to study professional narration quality while you refine your own production skills.

The settings are the foundation. Your voice is the building. Get both right, and you’ll have an audiobook that stands out for all the right reasons.