Why Technical Specifications Matter for ACX Narrators

Recording an audiobook for ACX (Audiobook Creation Exchange) is a creative endeavor, but it is also a technical one. Even the most compelling narration can be rejected if the audio fails to meet the platform’s rigorous technical standards. These specifications exist to ensure a consistent, professional listening experience across all devices and playback systems. For indie narrators and voice actors, mastering these details is the difference between a smooth submission process and repeated uploads, costly rework, or outright rejection.

ACX supplies these requirements to Audible, iTunes, and Amazon customers, and any deviation can lead to failed quality checks. This article expands on the core specs every narrator must internalize—and adds context on how to achieve them reliably in your home studio.

Core Audio Quality Requirements

ACX is explicit about the digital audio parameters your recordings must use. These are not optional suggestions—they are the baseline for acceptance.

Sample Rate and Bit Depth: Why 44.1 kHz / 16‑bit?

Record and export at a sample rate of 44.1 kHz with a bit depth of 16-bit. This standard is the same as CD-quality audio and is the most widely supported format across consumer playback devices. Why 44.1 kHz? It captures the full range of human hearing (20 Hz–20 kHz) without unnecessary data overhead. The 16-bit depth offers a dynamic range of about 96 dB, which is more than sufficient for spoken word when recorded at proper levels. Higher specs like 48 kHz or 24‑bit are unnecessary for speech and can actually cause compatibility issues if not converted correctly.

Be cautious: some recording software defaults to 48 kHz or 24‑bit. While those settings are fine for video or music production, they must be converted to the ACX specification before submission. Failing to do so will trigger a “wrong sample rate” rejection. To avoid this, set your session’s sample rate to 44.1 kHz before you record your first chapter. Use software that lets you lock the sample rate for the entire project, and check the export settings every time you render a final WAV file.

File Format: Lossless Delivery Is Non‑Negotiable

Export your final, mastered audiobook as a WAV or FLAC file. These are lossless formats that preserve the full audio quality. If your production workflow uses MP3 for review copies or chapter proofs, ensure the final deliverable is encoded at a constant bitrate of 192 kbps or higher—though FLAC or WAV is strongly preferred for final submission. MP3 is acceptable only for proofing; the final master must be lossless to avoid generation loss from repeated encoding.

ACX’s upload system will automatically convert your file to the appropriate format for distribution, but starting with a lossless source guarantees no generation loss. For more details, read the official ACX audio submission requirements.

Levels and Dynamics: Peak, RMS, and Headroom

Getting your recording level right is where many narrators stumble. ACX requires that your audio does not clip and maintains a consistent listening level.

Peak Level: The -3 dBFS Safety Margin

Your peak level must never exceed -3 dB (relative to full scale). Digital clipping occurs at 0 dBFS, so the -3 dB ceiling gives a safety margin. Use a limiter or careful gain staging to ensure no transient spike—like a loud plosive or sibilant “s”—breaches this limit. In practice, aim for peaks to hover around -6 dB during normal speech, then raise the overall gain so the loudest moment hits -3 dB. This reduces the workload on your limiter and helps avoid distortion.

RMS Level: The Sweet Spot for Spoken Word

RMS (Root Mean Square) measures the average loudness of your narration. ACX recommends an RMS level around -18 dB. This is typically considered the “sweet spot” for spoken word: loud enough to be heard clearly in a car or on a phone speaker, but not so loud that it causes listener fatigue. Most professional audio editing software includes a loudness meter or RMS analysis tool. Use it to scan your entire recording and make level adjustments per chapter. If you find your RMS is too low (e.g., -22 dB), apply compression or gain envelope to bring it up, but never at the expense of natural dynamics.

Why RMS Consistency Matters Across Chapters

Listeners will notice if one chapter is noticeably quieter or louder than the next. ACX’s quality reviewers also check for consistency. Aim to keep RMS within ±2 dB across all files. A common workaround is to use a compressor or leveler set to a low ratio (e.g., 2:1) with a threshold around -20 dB. This smooths out minor variations without sounding unnatural. After compression, fine‑tune each chapter’s gain so their integrated loudness matches. Tools like the YouLean Loudness Meter can give you precise LUFS readings, which correlate closely with RMS for speech.

Noise Floor, Room Acoustics, and Microphone Techniques

Technical specs aren’t just about numbers—they also concern the cleanliness of your audio signal.

Minimum Noise Floor: Inaudible Is the Goal

ACX does not publish a specific dB noise floor requirement, but the practical expectation is that background noise (fans, traffic, computer hum) should be inaudible during silent passages. A low noise floor, typically around -60 dB or lower, is desirable. Use a noise gate or noise reduction plugin if necessary, but be careful not to degrade the vocal quality. Spectral noise reduction (like iZotope RX’s Voice De‑Noise) can remove consistent hums without adding artifacts.

Room Acoustics: The Foundation of Clean Audio

Your recording space should be treated to minimize reverb, echo, and reverberation. Portable vocal booths, heavy curtains, and acoustic foam panels can help. Even a walk-in closet filled with clothes can work. The goal is a dry, neutral acoustic environment that lets the narrator’s voice be the star. Pay special attention to corners and flat parallel walls—these cause standing waves that color the sound. Bass traps in corners and rugs on hard floors reduce low‑frequency buildup.

Microphone Distance and Plosives

Position the microphone 6–12 inches from your mouth, slightly off-axis to reduce breath pops. Use a pop filter to tame plosives (P, B, T sounds). Plosives that exceed the -3 dB peak limit can cause rejection, and they are difficult to fix cleanly in post-production. Prevention is far easier. If you still get plosives, try a dynamic microphone like the Shure SM7B, which has built‑in pop filtration and a tighter pickup pattern.

Editing Workflow: Clicks, Breaths, and Cleanliness

Raw narration will always contain mouth clicks, lip smacks, and audibly harsh breaths. Editing these out is part of the narrator’s job.

Breath Removal and Mouth Click De‑clicking

Remove every breath that is louder than the natural flow of speech—especially deep inhales between sentences. Use a de‑clicker plugin (such as iZotope RX or Waves WLM Plus) to eliminate mouth clicks. Many editors also use a spectral frequency display to visually identify and cut clicks. The final audio should have a polished, professional sound with no distractions. For heavy mouth noise, try drinking water with a pinch of salt or using an apple slice to reduce dry mouth before recording.

Consistent Room Tone: Your Secret Weapon

Before you begin recording, capture at least 10 seconds of room tone (silence in your recording space). This helps when you need to fill gaps or apply noise reduction. It also allows you to check that your background noise is consistent across takes. Keep a short room‑tone clip in your editing project so you can paste it into any silent gaps—this prevents jarring audio dropouts when you cut out breaths.

File Naming, Chapter Markers, and Metadata

Technical specifications extend beyond the audio waveform. ACX requires correct file naming and embedded metadata for each chapter.

File Naming Convention

Each chapter file should be named with the book title and chapter number, for example: BookTitle_Chapter_01.wav. Do not use special characters or spaces. ACX provides guidelines in your title’s production dashboard. Following them prevents automated sorting errors. A common mistake is using underscores inconsistently (e.g., mixing hyphens and underscores). Stick to the pattern ACX shows in your dashboard for that specific project.

Chapter Markers and Part Labels

If your audiobook contains multiple parts or chapters, embed chapter markers using software like Audacity, TwistedWave, or Reaper. The markers define where listeners can skip or resume. ACX also requires a “Chapter Start” and “Chapter End” signal (a slight pause or fade). Check the ACX technical requirements for audiobook creation for the exact marker format. Also note: if your book has multiple parts (e.g., Part 1, Part 2), each part is treated as a separate “chapter” file from ACX’s perspective.

Technical Specifications Summary (Quick Reference)

  • Sample Rate: 44.1 kHz
  • Bit Depth: 16-bit
  • File Format: WAV or FLAC (MP3 minimum 192 kbps only for proofs)
  • Peak Level: -3 dB or lower (do not exceed)
  • RMS Level: -18 dB (aim for ±2 dB consistency)
  • Noise Floor: inaudible in silent passages
  • Breaths & Clicks: remove all audible non-speech sounds
  • File Naming: Title_Chapter_## (no spaces/special characters)
  • Chapter Markers: embedded in each WAV/FLAC file

Common Pitfalls and How to Avoid Them

Pitfall #1: Over‑EQ’ing the Voice

Many narrators try to “enhance” their voice with EQ—boosting lows or adding air. ACX prefers a natural, uncolored sound. A gentle high‑pass filter at 80 Hz to remove rumble is fine, but avoid drastic curves. Listen on headphones, monitors, and phone speakers to ensure your EQ doesn’t create harshness. If you feel your voice needs EQ, make small cuts rather than boosts—especially below 100 Hz and above 10 kHz.

Pitfall #2: Limiting Too Aggressively

To hit the -3 dB peak limit, some narrators push the limiter too hard, causing audible pumping or distortion. Instead, record at a lower level (peaks around -6 dB) and then apply gain staging to bring the RMS up to -18 dB, using a limiter only for the occasional transient. A gentle limiter with 2–3 dB of gain reduction is usually sufficient.

Pitfall #3: Forgetting to Check Mono vs. Stereo

ACX expects a mono file (one channel) unless your audiobook requires stereo effects (rare). Recording in mono reduces file size and prevents phase issues. Double-check your recording software’s track settings before you begin. If you accidentally record in stereo, collapse the two channels into mono by summing them (not just muting one side). Some software like Audacity has a “Split Stereo to Mono” option.

Pitfall #4: Skipping Room Treatment

Even the best microphone can’t compensate for a boxy, echo‑filled room. Don’t rely solely on noise reduction plugins; invest in acoustic treatment first. A $100 investment in moving blankets and PVC pipe can dramatically improve your sound.

Setting Up a Home Studio for ACX Compliance

You don’t need a multi‑thousand‑dollar studio. Here’s a realistic list:

  • Microphone: A large‑diaphragm condenser (e.g., Audio‑Technica AT2020, Rode NT1) or a dynamic mic (e.g., Shure SM58) in a quiet space. For narration, a condenser often provides richer detail, but a dynamic mic can handle poorer acoustic environments better.
  • Interface: A USB interface like Focusrite Scarlett 2i2 provides clean preamps. Ensure it supports 44.1 kHz at 16‑bit without interpolation.
  • Booth: A DIY booth using PVC pipes and moving blankets, or a purpose‑built iso booth. A simple carton box lined with acoustic foam can work for a portable solution.
  • Software: Audacity (free), Reaper ($60), or TwistedWave (subscription) all support ACX specifications. Reaper is especially popular for its flexible routing and batch processing.
  • Acoustic Treatment: Bass traps in corners, foam panels at first reflection points, and a rug on hard floors. Prioritize treatment at the reflection points—where sound bounces from the mic to your ears.

Final Quality Check Before Submission

Before you upload, perform a thorough quality control (QC) pass. Listen to every chapter in full at a moderate volume through closed‑back headphones. Check for:

  • Clipping or distortion
  • Mouth clicks and breaths
  • Inconsistent volume levels
  • Paper rustles, chair squeaks, or clothing noise
  • Missing chapter markers or incorrect file names
  • Audio drift (sync issues in multichapter projects)

Use ACX’s own checking tool—the ACX Checker app—to scan your files automatically. It catches many common errors like sample rate mismatches, peak overs, and insufficient headroom. You can download it for free from the ACX website. For a detailed walkthrough, refer to how to use the ACX Checker tool. Note that the checker runs on Windows and Mac, and it can be used offline for batch processing. Always run it on the final exported WAV files, not on your session files.

Conclusion: Precision Pays Off

Meeting ACX’s technical specifications is not about being a perfectionist—it’s about delivering a product that stands up to professional standards. Listeners pay for a seamless experience, and the platform enforces it. By mastering sample rates, levels, noise reduction, and file structure, you sidestep the most common rejection reasons and build a reputation for reliability. Every successful submission reinforces your credibility as a narrator who respects both the craft and the business of audiobook production.

Keep these specs handy, test your recordings early, and invest time in your editing process. The effort pays back in fewer revisions, faster approvals, and more time to do what you love: narrating stories. For additional resources, explore the Voiceover Herald’s ACX standards guide or the free ACX narration checklist from Audio Recording Services. Happy narrating!