Understanding ACX Noise Guidelines

ACX (Audiobook Creation Exchange), a platform connecting authors, narrators, and publishers, sets the standard for audiobook quality. To have your project approved and distributed through Audible, Amazon, and iTunes, it must meet specific technical specifications. Among the most critical are the requirements for audio levels and background noise. ACX mandates that the average RMS (root mean square) level of the recorded voice should be between -23 dB and -18 dB. This ensures consistent loudness across all chapters and tracks. Equally important is the noise floor, which must be at or below -60 dB RMS (A‑weighted). The noise floor is the level of ambient noise present when no voice is being recorded. Allowing background hum, room echo, or electronic interference to push the noise floor above -60 dB can lead to rejection. Listeners expect a clean, immersive experience free from distractions; meeting these technical requirements is the first step toward producing a professional audiobook that sells.

The Critical Role of Voice-to-Background Noise Balance

Balancing voice and background noise is not merely a technical checkbox—it directly affects listener comprehension and retention. When background noise is too high relative to the voice, listeners must strain to follow the narrative. This cognitive load causes fatigue and can drive away audiences. Conversely, a perfectly balanced track with a clear, forceful voice and a silent background allows the story to shine. Proper balance also ensures your submissions pass ACX’s automated audio review, avoiding costly re-recording or editing cycles. Mastering this balance demonstrates professionalism and an understanding of the listening environment—many consumers listen in cars, while commuting, or with earbuds, where background noise competitions are even more noticeable.

Step-by-Step Approach to Achieving the Ideal Balance

1. Create a Controlled Recording Environment

Start by tackling noise at the source. Record in a space that is as quiet and acoustically dampened as possible. A room with carpets, heavy curtains, and soft furniture will absorb reflections and reduce echo. For home studios, consider building a portable vocal booth or using heavy moving blankets to create a temporary isolation area. Turn off all unnecessary electronics (fans, air conditioners, refrigerators) and place your microphone away from computer vents and hard drives. Small investments in acoustic foam panels or a reflection filter can dramatically lower the noise floor before you ever hit record.

2. Invest in Quality Equipment

Your microphone and audio interface form the foundation of your recording chain. A large-diaphragm condenser microphone with a cardioid pickup pattern is ideal for voiceover work because it rejects sounds coming from the sides and rear. Avoid using built-in laptop microphones or cheap USB headsets—they capture excessive room noise and have limited dynamic range. Pair your microphone with a reputable interface that provides clean preamps and low self-noise. Use balanced XLR cables to minimize electrical interference. A good pop filter reduces plosive bursts, and a shock mount isolates the microphone from vibrations, which can travel through the floor or desk.

3. Apply Noise Reduction Strategically

Even in a quiet room, some residual noise remains. Use audio editing software like Audacity, Adobe Audition, or Reaper to apply noise reduction. The key is moderation: over‑aggressive noise reduction can produce “underwater” artifacts and degrade voice clarity. First, capture a few seconds of pure room tone (no speaking) to sample the noise profile. Then apply a noise reduction effect to the entire track, aiming to lower the noise floor to at least -65 dB while preserving the natural timbre of the voice. Always preview small sections to ensure you aren’t affecting voice quality. Noise gates can also be used during editing—they automatically mute the track when your voice falls below a certain threshold, effectively cutting out breaths, mouth clicks, and low-level hum between sentences.

4. Set Consistent Volume Levels

ACX’s average loudness range (-23 dB to -18 dB RMS) requires careful leveling. Use a compressor to reduce the dynamic range of your performance, making quiet parts louder and loud parts quieter. Start with a moderate ratio (2:1 or 3:1), slow attack, and medium release. After compression, normalize the entire track to an average of around -20 dB RMS as a safe target. Then adjust gain so that peak levels rarely exceed -3 dB to avoid clipping. Use a loudness meter plugin or tool to verify your RMS average. Many DAWs and standalone utilities (like Youlean Loudness Meter or the free TBProAudio dpMeter) can help you check both RMS and true peak levels.

5. Equalization (EQ) for Voice Clarity

EQ allows you to shape the frequency content of your recording. A typical spoken‑word EQ workflow includes a gentle high‑pass filter (around 80–100 Hz) to remove low‑frequency rumble from HVAC systems, footsteps, or structural vibrations. This alone can significantly clean up the noise floor. Next, a slight boost in the presence region (2 kHz–4 kHz) adds intelligibility and articulation. If sibilance (harsh “s” sounds) is an issue, notch cut around 6 kHz–8 kHz. Always make adjustments while listening on good headphones or studio monitors, and aim for a natural sound—over‑EQing can make the voice thin or “telephone‑like.”

6. Employ Compression for Consistency

Compression is your primary tool for maintaining even volume levels. Besides evening out performance dynamics, compression also brings up the perceived loudness of the voice relative to the background noise, making the noise less noticeable. For audiobooks, use a compressor with a threshold set so that it catches approximately 3–6 dB of gain reduction during the loudest parts of your reading. A faster attack (10–20 ms) helps control transients like hard consonants, while a slower release (50–100 ms) prevents “pumping” artifacts. After compression, a limiter can catch any remaining peaks and keep them below 0 dB to prevent distortion when the file is encoded.

Advanced Techniques for Professional Results

Noise Gates with Attack/Release Settings

A noise gate can be even more effective when tuned properly. Set the threshold high enough to close when you stop speaking but low enough to not cut off natural breath intakes. A fast attack (1–5 ms) ensures the gate opens instantly when you start speaking. A medium release (50–200 ms) lets the gate close gradually, avoiding an abrupt “chop” at the end of phrases. Use the gate as a secondary cleanup tool after noise reduction—do not rely on it solely, because it cannot remove noise present while you are speaking.

De‑Noising and Spectral Repair

For pesky noises that survive traditional filters (e.g., a distant dog bark, a car horn, or a door slam), spectral editing tools can surgically remove them. Programs like iZotope RX or Adobe Audition’s Spectral Frequency Display allow you to visually select the offending sound and eliminate it without damaging the voice. This is a last resort for severe issues, but with practice it can save a take that would otherwise have to be re‑recorded. Always back up your raw files before applying any destructive processing.

Monitoring with the Right Environment

If you have studio monitors, position them at ear level in an equilateral triangle with your listening position. For most home setups, a good pair of closed‑back headphones (like the Sony MDR‑7506 or Beyerdynamic DT 770 Pro) will give you a more accurate representation of the noise floor and balance. Avoid consumer earbuds or open‑back headphones, which can mask low‑level noise or leak audio back into your recording.

Practical Tips for Daily Recording Sessions

Develop a pre‑recording routine that includes checking for new sources of interference. Cell phones, Wi‑Fi routers, and even LED light bulbs can introduce high‑frequency noise. Before each session, turn off or move anything that might add electrical hum. Monitor your input level in your DAW—keep the preamp gain high enough to get a good signal‑to‑noise ratio (peaking around -12 dB to -6 dB) but not so high that you introduce clipping or self‑noise from the interface.

Listen critically to your raw recordings. Play back a full minute of your performance on loop while reading the script. Mark any sections where you hear clicks, pops, or sibilance. Fix these in editing rather than hoping they will go unnoticed. Use a de‑esser plugin to tame harsh “s,” “z,” and “sh” sounds—a dedicated tool is often more effective than EQ alone. Keep your recording environment consistent: if you down blankets or put away acoustic treatment between sessions, the noise signature will change, requiring new noise reduction settings for every chapter.

Take breaks every 45–60 minutes to rest your ears. Ear fatigue leads to poor editing decisions—you might miss a loud click or think the noise floor is quieter than it actually is. Hydrate well and avoid dairy products before recording, as they can cause mouth noises that later require manual cleanup.

Final Checks Before Submission

When you believe your file is ready, perform a systematic quality control checklist. First, listen to the entire file from start to finish—not just random samples. Use a loudness meter to confirm that the average RMS sits between -23 dB and -18 dB, and that the true peak does not exceed -3 dB. Verify the noise floor by selecting a silent passage of at least one second and reading the RMS level; it should be below -60 dB. If your meters show anything above -58 dB, go back and apply more noise reduction or re‑record the section.

Listen for any pumping artifacts from over‑compression, distortion from clipping, or unnatural silences caused by an overly aggressive noise gate. Check that the first and last 500 milliseconds of the file are silent (no clicks or pops at the head). ACX also requires that each track begins and ends with silence - no leading or trailing noise. Use an audio editor to fade in and fade out over 200‑300 ms at the very start and end of the file. Finally, export at the required sample rate (44.1 kHz) and bit depth (16‑bit), using constant bitrate MP3 at 192 kbps or the specified WAV format. Run the file through ACX’s free Checker tool (ACX Audio Submission Guidelines) to catch any technical violations before you upload.

If you are still struggling with persistent background noise, consider investing in a software suite like iZotope RX Elements (iZotope RX) which offers industry‑standard noise reduction and spectral repair. For a free alternative, Audacity (Audacity) provides solid noise reduction and compression tools—just be sure to learn their proper use through tutorials. Many narrators also share their settings and workflows on forums like the ACX Narrators group, which can be a goldmine of practical advice.

By systematically controlling your recording environment, applying targeted processing, and verifying every parameter against ACX standards, you will not only pass review consistently but also deliver a listening experience that earns five‑star ratings and loyal fans. The effort you invest in balancing voice and background noise directly translates into professional credibility and long‑term success in the competitive audiobook market.