Why ACX Compliance Matters for Audiobook Success

Publishing an audiobook through Audible, iTunes, or other major platforms means your audio must pass ACX's strict technical requirements. ACX (Audio Creation Exchange) is Amazon's platform for audiobook production and distribution, and it enforces specific standards for audio quality. If your files don't meet these thresholds, your submission gets rejected, costing you time and money. Understanding the complete editing workflow is not just optional — it is essential to getting your audiobook onto store shelves.

Before you can edit effectively, you need to know exactly what ACX checks. The official ACX audio submission requirements cover several key areas: loudness levels must fall between -23dB and -18dB RMS, with a maximum peak of -3dB. The noise floor should be at or below -60dB. Audio must be delivered as 192 kbps or higher MP3 files, or WAV files at 44.1 kHz sample rate with 16-bit depth. There should be no audible background noise, no clipping, no DC offset, and no excessive sibilance or plosives. Headroom must be sufficient to avoid distortion. These requirements are non-negotiable, and your editing workflow must address each one.

Many producers focus only on loudness and noise, but ACX also checks for consistent sound quality across all chapters, proper chapter markers, and metadata accuracy. The editing techniques below cover every aspect of ACX compliance, from the recording environment to final quality control.

Recording Environment and Capture: The Foundation of Clean Audio

No amount of editing can fully salvage a poorly recorded track. The most efficient path to ACX compliance starts before you hit record. Your recording space must be quiet, with minimal reverb and no external noise sources. A dedicated vocal booth, a closet filled with soft materials, or a room treated with acoustic panels and bass traps works well. The goal is to achieve a noise floor below -60dB without relying on heavy noise reduction later.

Microphone technique is equally critical. Position the microphone 6 to 12 inches from the narrator's mouth, off-axis from the nose to reduce plosive blasts. Use a pop filter at all times. Set your recording level so that the loudest peaks hit around -6dB to -3dB on your meter, leaving headroom for processing. Record at 24-bit depth and 48 kHz sample rate, then convert to 16-bit, 44.1 kHz for final delivery. This gives you more headroom and resolution during editing.

Record in short takes of 15 to 30 minutes to avoid fatigue-related errors. Leave 2 to 3 seconds of silence at the beginning and end of each take for noise profiling. This silence becomes your noise print for noise reduction later. Consistent recording levels and microphone positioning across all sessions prevent jarring shifts in volume or tone between chapters.

Essential Editing Techniques for ACX-Compliant Audio

Once your raw recording is captured, the editing process begins. Below are the techniques that directly address ACX requirements, ordered in a practical workflow.

1. Noise Reduction: Removing the Room from the Voice

Background noise is the most common reason for ACX rejection. Start by selecting a section of room tone or silence from your recording — ideally 2 to 3 seconds of pure background noise. Use a spectral noise reduction tool like iZotope RX, Adobe Audition, or Audacity's noise reduction function. Capture this as your noise profile. Apply noise reduction gently, aiming to reduce the noise floor to -60dB or lower without introducing artifacts like "warble" or "swimming."

A typical reduction of 12 to 18dB is sufficient for most home studios. If you need more than 24dB of reduction, your recording environment needs improvement. Over-reduction creates a hollow, unnatural sound that can also trigger ACX automated quality checks. Always listen critically after applying noise reduction. Use the "listen to noise only" mode to ensure you are not removing voice frequencies. For persistent low-frequency hum (60Hz or 50Hz), use a notch filter rather than broadband noise reduction.

2. Equalization (EQ): Shaping the Voice for Clarity and Warmth

EQ fine-tunes the frequency balance of your narration. The human voice ranges from roughly 80Hz to 8kHz, but audiobook narration benefits from a focused, clear midrange. Start with a gentle high-pass filter at 60-80Hz to remove subsonic rumble and handling noise. This immediately cleans up the low end and helps meet the -60dB noise floor requirement.

For the main EQ curve, consider a slight boost around 3kHz to 5kHz for presence and intelligibility. A small cut around 200Hz to 400Hz can reduce muddiness. Avoid boosting below 100Hz; that range adds boominess without improving speech clarity. If the narrator's voice sounds thin, a gentle shelf boost above 5kHz can add air. Always apply EQ in small increments — 2 to 3dB maximum. Over-EQing introduces phase issues and unnatural coloration. Use a linear-phase EQ for surgical cuts and a minimum-phase EQ for broader tonal shaping. Check your EQ changes on both studio monitors and headphones to ensure consistency across playback systems.

3. Compression: Controlling Dynamics for Consistent Loudness

Audiobook audio needs consistent volume levels to avoid listener fatigue. Compression narrows the dynamic range so that quiet passages remain audible without loud peaks distorting. Set your compressor with a ratio of 2:1 to 4:1, a fast attack time (1 to 10 milliseconds), and a medium release time (50 to 100 milliseconds). Adjust the threshold so that the compressor reduces gain by 3 to 6dB on average peaks. The goal is not to squash the life out of the performance but to create a smooth, even level.

After compression, use a makeup gain stage to bring the overall level up. Monitor your RMS or LUFS level. ACX requires -18dB to -23dB RMS. Use a loudness meter — like the free YouLean Loudness Meter or the built-in meters in your DAW — to verify your integrated loudness. Aim for -20dB RMS as a safe target. Avoid heavy limiting to achieve loudness; limiting should only catch occasional peaks above -3dB. If you find yourself applying more than 6dB of compression, consider either re-recording with better microphone technique or using a multiband compressor for more precise control.

4. De-essing and Plosive Reduction: Eliminating Harshness and Pops

Excessive sibilance — the harsh "s," "sh," "ch," and "z" sounds — is a common compliance issue. Use a dedicated de-esser plugin or a dynamic EQ with a narrow band around 5kHz to 8kHz. Set the threshold so that only the sibilant consonants trigger the reduction, typically 4 to 6dB of gain reduction. Avoid over-de-essing, which creates a lisping effect. Some de-essers offer a split-band mode that targets only the high frequencies, preserving the body of the voice.

Plosives — the explosive "p," "b," "t," and "k" sounds — create low-frequency bursts that can distort and fail ACX peak limits. If your pop filter and microphone positioning did not eliminate them, use a spectral repair tool to manually remove or reduce the plosive burst. Alternatively, draw a volume automation curve to dip the plosive by 6 to 10dB over 50 to 100 milliseconds. Do not simply clip or delete the waveform; that creates an audible gap. Surgical editing ensures the narration remains smooth and natural.

Advanced Editing Techniques for Polished Narration

Once the core processing is done, advanced techniques elevate your audio from compliant to exceptional. These address subtle issues that automated checks might miss but listeners will notice.

Mouth Click Removal and Breath Management

Mouth clicks, lip smacks, and other small noises are common in long narration sessions. Use a spectral editing tool to visually identify and remove these clicks. iZotope RX's Mouth De-click tool is industry standard, but manual removal with a pencil tool works for simple clicks. Zoom in to the waveform and replace the click with a tiny silence or a crossfade from the surrounding audio. Removing excessive breaths also improves pacing. Reduce breath volume by 50% to 75% rather than removing them entirely; complete silence where a breath should be sounds unnatural. Try a breath management plugin like Waves NS1 or iZotope Breath Control to automate this process.

Chapter Alignment and Crossfade Consistency

ACX requires each chapter to be a separate file with consistent volume and tonal quality. After editing individual chapters, audition them in sequence to ensure there are no jumps in volume or frequency balance. Use a loudness match or gain normalization tool to bring all chapters to the same RMS level, usually within 0.5dB of each other. Apply a subtle 10 to 20 millisecond crossfade at the beginning and end of each chapter to prevent clicks during playback. Label each file with the correct chapter number and title per ACX metadata guidelines.

Room Tone Matching for Seamless Edits

When you edit out mistakes or long pauses, you create cuts in the audio. If the room tone changes between takes, those cuts become audible. Create a loop of clean room tone from your recording environment. Use this loop to fill any gaps created by editing. Match the level of the room tone to the existing background noise by ear or using a spectrum analyzer. This technique ensures that every edit sounds like a continuous, uninterrupted performance. It is especially important for audiobooks recorded across multiple sessions.

Mastering and Final Preparation for ACX Submission

The final stage is mastering your processed audio into the exact file format ACX expects. Do not skip this step; improper file formatting is a frequent cause of rejection.

Loudness and Peak Verification

Run your entire file through a loudness meter one last time. ACX uses RMS loudness, not LUFS, though many meters display both. Your average RMS should be between -18dB and -23dB. The maximum peak should never exceed -3dB. If your file is too quiet, apply gain staging or gentle limiting. If it is too loud, reduce the makeup gain from your compressor or lower the overall track level. Use a true peak limiter with a ceiling of -3.5dB to catch any intersample peaks that could clip after encoding.

File Format and Sample Rate Conversion

Export your final audio as a mono WAV file at 44.1 kHz sample rate and 16-bit depth. ACX accepts stereo files, but mono is standard for audiobooks and reduces file size. If you recorded at 48 kHz, you must convert down to 44.1 kHz using a high-quality sample rate converter (SRC). Dithering from 24-bit to 16-bit needs a noise-shaped dither to preserve dynamic range. Use the built-in dithering in your DAW or a dedicated plugin like iZotope MBIT+.

Metadata and Chapter Markers

ACX requires chapter markers in your audio file. In your DAW, place markers at the start of each chapter with the chapter title and number. Export these markers using the appropriate format (CUE, markers track, or embedded metadata). Verify that the total number of chapters matches your metadata submission. Check that the track title, artist name, and publisher fields are correctly filled. Incorrect metadata causes automated rejection even if the audio is perfect.

Quality Control: The Safety Net Before Submission

Quality control (QC) is the final, critical step. Do not submit immediately after exporting. Let your ears rest for 24 hours if possible. Listen to the entire audiobook from start to finish, preferably on multiple playback systems — headphones, earbuds, laptop speakers, and a car stereo. Listen for:

  • Volume inconsistencies between chapters or within a chapter
  • Audible edits or unnatural pauses
  • Background noise that changes between cuts
  • Distortion, clipping, or pumping from compression
  • Sibilance or plosives that escaped earlier processing
  • Mouth clicks or breath artifacts that break immersion

Consider using automated QC tools like Auphonic or ACX Check to pre-verify your loudness and noise floor. But never rely solely on software. Human hearing catches issues that meters miss, like a slight tonal shift or an emotional disconnect in the narration. If possible, send your master to a second producer or a QC service for a fresh perspective. A second set of ears regularly catches errors you have become deaf to through repeated listening.

Finally, run your file through ACX's own submission checker. Upload a sample chapter to the ACX website and use their built-in validation tool. It will flag loudness errors, clipping, noise issues, and formatting problems. Fix any issues before uploading the full project. A rejected submission delays your release and damages your reputation on the platform.

Common ACX Rejection Reasons and How to Avoid Them

Understanding why files get rejected helps you preemptively fix problems. The most common rejection reasons include:

  • Noise floor too high (above -60dB) — fix with better recording environment or gentle noise reduction
    • Loudness outside -18 to -23dB RMS — adjust compression and makeup gain
  • Peak level exceeds -3dB — use a true peak limiter
  • Audible background noise changes — ensure room tone matching
  • Mouth clicks and breaths too prominent — use spectral repair and breath reduction
  • Sample rate or bit depth incorrect — check your export settings

Build a checklist from these rejection points and go through it before every submission. Over time, the process becomes routine, and rejections become rare.

Conclusion: Building a Reliable Editing Workflow for ACX Success

Achieving ACX-compliant audiobook audio is a repeatable process built on solid recording practices, precise editing techniques, and thorough quality control. Start with a clean capture environment and microphone technique. Apply noise reduction, EQ, compression, de-essing, and plosive reduction in a logical order. Use spectral tools for surgical repairs like mouth clicks. Master your audio to the exact ACX format specifications. Then verify everything with your own ears and automated tools.

Consistency across chapters and sessions sets professional work apart from amateur submissions. Develop templates in your DAW with your preferred settings for EQ, compression, and loudness metering. Save presets for noise reduction and de-essing tailored to your microphone and voice. Every narrator's voice and room are different, so fine-tune these presets over time. The result is a faster, more reliable workflow that produces clean, compliant audio every time.

For further reading, consult the official ACX Audio Specifications page for the latest requirements. Community resources like the Audiobook Creation Exchange resource hub offer practical tips. For advanced audio post-production techniques, tutorials from iZotope's audiobook processing guide provide detailed walkthroughs. With a methodical approach, you can produce audiobooks that pass ACX checks easily and keep listeners engaged for hours.