Understanding ACX Sound Quality Standards

Entering the world of audiobook narration is an exciting step, but your audio must meet the rigorous technical standards set by ACX (Audiobook Creation Exchange). ACX acts as the gateway to major retailers like Audible and iTunes, and they enforce strict quality checks to ensure a consistent listening experience. Missing these specs means your files will be rejected, costing you time and effort. This guide walks through each requirement, how to measure them, and the equipment and techniques you need to pass ACX validation on your first submission.

The Six Critical ACX Audio Specifications

ACX publishes a clear set of requirements. Every file you upload must conform to these exact parameters. Below is a breakdown of each spec and why it matters.

  • Sample Rate: 44.1 kHz – This is the standard for CD-quality audio. It captures the full range of human hearing (up to 22 kHz) without introducing unnecessary data. Recording at 48 kHz or 96 kHz and downsampling is acceptable, but always export at 44.1 kHz for ACX. Downsampling incorrectly can introduce aliasing artifacts.
  • Bit Depth: 16-bit – 16-bit depth provides 96 dB of dynamic range, more than enough for spoken word. Recording in 24-bit gives you more headroom during editing, but you must dither down to 16-bit for the final export. Never submit 24-bit files; ACX will reject them.
  • Audio Format: Mono or Stereo – Most audiobooks are mono, because narration comes from a single point source. Stereo files double the file size with no benefit and can cause phase issues if the listener uses headphones. ACX accepts both, but mono is strongly recommended. Check your DAW's export settings to ensure mono interleaved or joint stereo is not accidentally selected.
  • Noise Floor: Maximum -60 dB (unweighted) – This is the most common reason for rejection. The noise floor is the level of background hiss, hum, or room tone present during silences. ACX measures this with an RMS meter. You must achieve a consistent floor of -60 dBFS or lower. With good recording technique and proper noise reduction, this is attainable even with budget gear.
  • Signal-to-Noise Ratio: At least 35 dB – This ensures that the voice is clearly above the noise floor. ACX requires the average RMS level of your narration to be between -18 dBFS and -23 dBFS. Combined with a -60 dBFS floor, you get a healthy 37+ dB separation.
  • Peak Level: No higher than -3 dBFS – Digital clipping occurs when peaks hit 0 dBFS. ACX demands headroom, so your loudest peaks (shouts, plosives, sibilance) must stay at -3 dB or lower. Use a limiter or manual gain staging to prevent overs.

The official ACX audio submission requirements page is your definitive reference. Bookmark it and check back for updates before each submission.

Setting Up Your Recording Environment for ACX Compliance

Your room is the single largest factor affecting audio quality. A cheap microphone in a well-treated room will outperform an expensive mic in a reflective, noisy space. Focus on two goals: eliminating external noise (traffic, HVAC, appliances) and controlling reflections (echo, reverb).

Budget-Friendly Acoustic Treatment

You don't need a professional recording booth. Many successful narrators use portable vocal booths or DIY panels. Key strategies include:

  • Deadening the space: Use heavy blankets, moving blankets, or acoustic foam panels on walls, especially behind and to the sides of the microphone. The goal is to absorb high and mid frequencies. Avoid leaving bare parallel walls that cause flutter echo.
  • Isolating your mic: A reflection filter (like a sE Electronics RF-X) placed behind the mic helps in mildly treated rooms. For better results, build a small gobo (portable absorber) around your recording area.
  • Eliminating bass frequencies: Low-end rumble from HVAC or traffic is harder to remove. A high-pass filter (around 80 Hz) during recording or editing can clean up the track without affecting voice quality.

Essential Equipment for Clean Recordings

You can achieve ACX-compliant audio with a modest budget if you choose wisely. Below is a checklist with practical recommendations.

  • Microphone: A large-diaphragm condenser (e.g., Audio-Technica AT2020, Rode NT1-A) offers sensitivity and detail, but requires a quiet room. A dynamic mic (e.g., Shure SM58, Electro-Voice RE20) picks up less room noise and is more forgiving for untreated spaces. Both types can pass ACX checks when used correctly.
  • Audio Interface: Avoid USB microphones if possible; they often have noisy preamps and limited control. A dedicated interface (Focusrite Scarlett 2i2, Universal Audio Volt 2) provides clean gain and low noise floor. Set the gain so you average around -18 dBFS on your DAW's meters.
  • Pop Filter & Boom Arm: A double-layer pop filter stops plosive blasts that cause low-frequency thumps (clipping). Pair it with a sturdy boom arm to position the mic off-axis (slightly above or to the side) to reduce sibilance.
  • Closed-Back Headphones: Open-back headphones leak sound into the mic. Use closed-back models (Audio-Technica ATH-M50x, Sony MDR-7506) for monitoring while recording. You can switch to open-back for editing.

For a deep dive on microphone placement and room treatment, read Sweetwater's guide to home studio acoustic basics.

Recording Techniques That Prevent Common Pitfalls

Even with perfect gear, poor microphone technique can ruin your take. Implement these habits from day one.

Consistent Mic Position and Distance

Maintain a distance of 6–12 inches from the microphone diaphragm. Closer increases proximity effect (boomy bass), which can push you below -23 dBFS average when you back off. Farther away picks up more room tone and reduces signal strength, forcing you to add gain that amplifies noise. Use a pop filter as a physical reference; keep your lips about two finger-widths from the filter.

Mouth Clicks and Sibilance Prevention

Mouth noises are a major source of ACX rejections because they become obvious during silences. Stay hydrated (water, not sugary drinks) and avoid dairy products before recording. Use a de-esser in post to tame sibilance (typically in the 5–8 kHz range). For persistent clicks, try a specialized plugin like iZotope RX Mouth De-click.

Room Tone Capture for Noise Reduction

Before each session, record 10–15 seconds of silence in your recording space (no speaking, no movement). This "room tone" becomes your noise profile. You will use it during editing to apply noise reduction precisely without harming the voice. Without this, your noise reduction may sound unnatural.

Post-Production: Achieving ACX-Compliant Audio in Your DAW

Editing is where you transform a raw recording into a polished, standards-compliant file. The order of processing matters. Follow this chain to minimize artifacts and maximize clarity.

Step 1: Noise Reduction

Use your room tone capture to create a noise print. In Audacity, select the room tone clip, go to Effect > Noise Reduction > Get Noise Profile. Then select the entire track, open Noise Reduction again, and reduce noise by 12–20 dB (start with 12 dB to avoid "underwater" artifacts). In Reaper or Pro Tools, similar plugins exist. Always use a high-pass filter before noise reduction to avoid processing low-frequency rumble that can distort the algorithm.

Step 2: Compression for Consistent Levels

ACX requires an average RMS of -18 to -23 dBFS. A compressor evens out loud and quiet passages. Set a moderate ratio (2:1 to 3:1) with a threshold around -20 dBFS. Aim for about 3–6 dB of gain reduction on peaks. Use a fast attack (10–20 ms) and a medium release (50–100 ms) to keep the voice natural. Follow compression with gain makeup to bring the level up to -18 dBFS average.

Step 3: Equalization (EQ)

A light EQ can enhance clarity and remove problematic frequencies. Typical adjustments:

  • High-pass filter at 80 Hz (removes low-end rumble)
  • Moderate 2–3 dB boost at 3–5 kHz (adds presence and intelligibility)
  • Gentle cut at 200–400 Hz if the voice sounds muddy
  • De-esser placed after compression to handle sibilant spikes

Step 4: Normalization and Limiting

After compression and EQ, normalize the track to -1 dB peak (or use a brickwall limiter set to -3 dBFS to ensure you never exceed ACX's peak limit). Then use a loudness analyzer to check the average RMS. Adjust the overall gain until the RMS reads between -18 and -23 dBFS. For Audacity, use the "Loudness Normalization" effect set to -19 dBFS (a safe midpoint).

Final Quality Control: The ACX Check and Beyond

Before uploading, run your file through the ACX Check tool. This free online analyzer simulates ACX's validation process. Upload your WAV or MP3 and it reports sample rate, bit depth, peak level, noise floor, and RMS. Many narrators also recommend Auphonic, which automatically adjusts levels to ACX standards and can process multiple files in batch. However, reviewing manually ensures you understand what's happening to your audio.

Common Rejection Reasons and Fixes

  • Noise floor too high: Re-record with better isolation, or apply more aggressive noise reduction. Check that your mic preamp gain isn't too high (cranking gain increases noise).
  • Volume too quiet or too loud: Use an RMS meter. If your average is below -23 dBFS, apply makeup gain; if above -18 dBFS, reduce gain via volume automation or compression.
  • Clipping: Look for red waveforms. Use a limiter (set to -3 dBFS) or manually lower the gain on those sections. If clipping exists in the raw recording, re-record those lines.
  • Background noise during pauses: Edit out all breaths and mouth clicks in silence (use spectral editing). If the noise is constant, use noise reduction. Never leave empty space with any audible room tone.

A helpful resource for advanced editing techniques is the iZotope guide to noise reduction for audiobooks. It covers spectral repair, matching EQ, and loudness optimization.

Submitting Your Audiobook to ACX

When your files pass the ACX Check, export them as either WAV (16-bit, 44.1 kHz, mono) or MP3 (320 kbps, constant bitrate). ACX prefers WAV for the highest quality, but both are accepted. Follow ACX's file naming convention: AuthorName_BookTitle_PartNumber.wav. Upload each part individually in your ACX project dashboard. Ensure your metadata (title, author, narrator, ISRC codes) is correct. After submission, ACX will run its own automated validation; if it fails, they'll provide a reason. Most issues are correctable without re-recording the entire book.

Meeting ACX standards is a skill that improves with practice. Focus on your recording environment first, then refine your editing workflow. Many narrators produce professional audiobooks with less than $500 in equipment. The key is meticulous attention to each spec and consistent quality checks. With this guide and the resources linked throughout, you are equipped to produce ACX-compliant audiobooks that listeners will enjoy.