Audiobook Creation Tools Settings — A Practical Guide

You don’t need a recording studio to produce a professional-sounding audiobook. The gap between amateur and polished narration often comes down to a handful of settings you can adjust in free or affordable software. This guide walks through the essential controls, the order to touch them, and what each one actually does to your voice.

Start With Your Recording Space, Not Your Software

Before you open any tool, your room is the first “setting” to fix. A quiet room with soft furnishings—curtains, a carpet, a bed—absorbs the echo that makes amateur recordings sound hollow. Closets full of clothes work surprisingly well as makeshift vocal booths.

For the microphone itself, keep the gain low. Most USB mics (like the Audio-Technica ATR2100x or Samson Q2U) sound best with gain set to around 50–60% of maximum. This leaves headroom for loud passages without clipping. You can always boost volume later; you cannot un-distort a clipped recording.

A quick check before you record: do a 30-second test take and watch the waveform. If the peaks are touching the top and bottom edges of the track, your gain is too high. If the waveform looks like a thin ribbon hovering near the center line, it’s too low—you’ll introduce noise when you boost it later. Aim for peaks that reach about halfway up the track. If you see flat-topped peaks that look chopped off, that’s clipping; lower the gain immediately and re-record. No amount of editing can repair a clipped waveform.

The Core Settings in Audacity (Free and Reliable)

Audacity remains the default starting point for most self-produced audiobook narrators. It is free, runs on any computer, and handles the essential edits without fuss.

Set your project rate to 48,000 Hz and record in mono. Audiobooks are mono by industry standard—your voice comes from one speaker, and stereo files just double the file size for no benefit.

When you open the recording for editing, work in this order:

1. Noise Reduction — Capture a few seconds of pure room tone (silence with the room’s ambient hum), then use Audacity’s Noise Reduction effect to sample that profile and remove it from the entire track. Overuse creates a watery, underwater sound, so keep the reduction amount around 10–15 dB.

2. Normalize to -3 dB — This sets your loudest peak to a consistent level. Audiobook platforms expect peaks around -3 dB to leave room for playback systems.

3. Compressor (Light Touch) — A 2:1 ratio with a -12 dB threshold smooths out volume swings between quiet and loud sentences. You are not trying to squash dynamics; you are preventing listeners from constantly adjusting their volume.

4. Loudness Normalize to -16 LUFS — This is the target for spoken-word content on most platforms. Audacity’s Loudness Normalization effect handles this in one pass.

A concrete example: the audiobook narrator Steven Pacey, who reads Joe Abercrombie’s The First Law series, is famous for dynamic range—he whispers, growls, and shouts within a single chapter. A heavy-handed compressor would flatten exactly what makes his narration gripping. Use compression to tame peaks, not to erase expression.

Here’s where the branch happens. After you run noise reduction, listen to the first 30 seconds of your track. If you hear a watery, chorus-like artifact—as if your voice is slightly underwater—you’ve over-applied the effect. In that case, undo the noise reduction, re-sample your room tone, and reduce the reduction amount to 6–8 dB. You can also try applying noise reduction in two lighter passes instead of one heavy pass.

If the background hum is still audible after a second attempt at lighter settings, your room tone is too loud to fix in software; stop editing and address the space itself—close windows, unplug appliances, or move to a quieter room. A persistent hiss or electrical buzz that survives noise reduction is a signal to check your cables and USB port before continuing.

Reaper: The Next Step Up for Serious Producers

If you outgrow Audacity, Reaper costs $60 for a personal license and offers far more control. The key settings to adjust here:

• Sample rate: 48 kHz, 24-bit depth — The extra bit depth gives you more headroom when editing quiet passages.
• Monitoring FX chain — Set up a light compressor and EQ on your monitoring path (what you hear while recording) without baking those effects into the file. This lets you hear a polished version while recording raw audio for maximum flexibility later.
• Render settings — Export as WAV at 48 kHz, then convert to MP3 at 320 kbps for distribution. Audible accepts both, but WAV is the master copy you should archive.

Reaper’s learning curve is steeper than Audacity’s, but its batch processing and project templates save hours if you record multiple chapters per session.

A practical workflow for multi-chapter projects: record all chapters in one session if possible. If you must record across multiple days, match your mic position exactly—same distance from your mouth, same angle. Use a pop filter as a physical reference point. When you render, process every chapter with the identical chain and verify the loudness reading on each file. If Chapter 3 comes out at -15.2 LUFS and Chapter 4 at -16.8 LUFS, your ears will notice the jump even if the numbers seem close. Re-run loudness normalization on the outlier chapter until all files sit within 0.5 LUFS of each other.

EQ: The Setting Most Beginners Skip

Equalization is where amateur narrators lose listeners without knowing why. Most USB mics add a slight muddiness around 200–300 Hz and a harshness around 3–5 kHz.

A simple corrective EQ curve for narration:

• High-pass filter at 80 Hz — Removes rumble from air conditioners, traffic, and mic handling.
• Cut 2–3 dB at 300 Hz — Reduces muddiness and improves clarity.
• Boost 2 dB at 3 kHz — Adds presence and intelligibility, especially at 1.5x playback speed, which your target listener likely uses.

The narrator Edoardo Ballerini, who has read everything from The Count of Monte Cristo to contemporary fiction, has a naturally warm voice. A heavy low-mid boost would turn that warmth into mud. The goal is a neutral, natural sound—not a “radio voice” effect.

How to verify your EQ is working: after applying the curve, listen to a passage with hard consonant sounds—words like “crisp,” “track,” “static.” If the S and T sounds feel harsh or spitty, reduce the 3 kHz boost to 1 dB. If your voice sounds thin or nasal, check whether the 300 Hz cut is too aggressive; dial it back to 1.5 dB. The test is simple: your voice should sound like you, just clearer—not like a different person.

Export Settings That Match Platform Requirements

Each audiobook platform has specific technical requirements, and getting these wrong means rejected files.

Audible (ACX) expects:

• MP3 at 192 kbps or higher, or WAV at 44.1/48 kHz
• Mono, not stereo
• -23 dB RMS with -3 dB peaks (their “loudness” standard)
• No background noise, clicks, or mouth sounds

Spotify and Apple Books are more flexible but generally accept the same ACX standards. If you produce to ACX specs, you can upload almost anywhere.

A quick check: after exporting, listen on earbuds, not just speakers. Earbuds reveal mouth clicks and sibilance that speakers mask. Fix these with a de-clicker (Audacity has a basic one; iZotope RX is the professional standard) and a de-esser set to around 6–8 kHz.

The verification step that catches most problems: before uploading, play your exported file on three different devices—your phone speaker, earbuds, and a car stereo if you have one. Each playback system reveals different flaws. Phone speakers exaggerate mid-range harshness; earbuds expose sibilance and mouth clicks; car stereos reveal bass rumble and room echo. If the file sounds clean on all three, you’re ready to submit. If one device exposes a problem you didn’t hear on the others, go back to the specific fix—de-esser for sibilance, high-pass filter for rumble—and re-export.

The 1.5x Speed Test

Your audience listens at accelerated speeds. A 2023 study from Journal of Experimental Psychology found that comprehension remains high up to 2x speed, but listener preference clusters around 1.5x. This changes how you evaluate your settings.

At 1.5x, every pause becomes shorter, every sibilant becomes sharper, and every room echo becomes more noticeable. Test your exported file at 1.5x before finalizing. If the narration sounds rushed or harsh at that speed, your pacing (not your settings) needs adjustment—longer pauses between sentences, slightly slower delivery.

Narrators like Julia Whelan, who reads The Invisible Life of Addie LaRue, pace their delivery with natural rhythm that survives speed-up. If you listen to her at 1.5x, the narration still breathes. That is the benchmark to aim for.

A specific test you can run right now: pick a 60-second passage from your recording. Listen to it at 1.0x, then at 1.5x, then at 1.25x. At 1.5x, pay attention to whether words blur together at the ends of sentences. If they do, your pauses between sentences are too short—extend them by 0.2–0.3 seconds in editing. If sibilant sounds (S, SH, CH) become piercing at 1.5x, your de-esser threshold needs to be more aggressive. The goal is a file that sounds natural at 1.5x without any single element—pacing, sibilance, or echo—standing out as a problem.

Common Settings Mistakes and How to Fix Them

Mistake 1: Over-processing. Applying noise reduction, compression, EQ, and limiting in sequence often produces a sterile, lifeless voice. Each effect removes something. Apply only what the recording needs, and trust your ears over a preset chain.

Mistake 2: Inconsistent loudness across chapters. If you record on different days, your voice and mic position shift. Use the Loudness Normalize effect on every chapter with the same target (-16 LUFS) to keep the listening experience consistent.

Mistake 3: Ignoring the room. No software setting can fix a recording made in a tiled bathroom. The echo is baked into the file. Fix the space first, then the settings.

Mistake 4: Skipping the listen-back. After editing, listen to the entire chapter in one sitting. Do not edit while listening; just listen. Your brain will catch pacing issues, repeated words, and awkward cuts that you miss when you are zoomed into waveforms.

When to stop and escalate: if you’ve gone through the full chain—noise reduction, normalization, compression, loudness, EQ—and your recording still has audible room echo, persistent hum, or a buzzing sound that changes when you touch the mic cable, stop editing. These are hardware or environment problems, not settings problems. Continuing to apply software fixes will degrade your voice quality while the underlying issue remains. At this point, check your cable connections, try a different USB port, or record a test in a different room. If the problem persists across rooms and cables, your microphone itself may be faulty—contact the manufacturer’s support or consider a replacement. A recording that requires more than two rounds of noise reduction to sound clean is not worth salvaging; re-record in a better space.

Building a Settings Template for Consistency

Once you find settings that work for your voice, save them as a template. In Audacity, this means saving a project with your effects chain applied to a blank track. In Reaper, save a project template with your FX chain, routing, and render settings pre-configured.

This matters because consistency across chapters is what separates a professional audiobook from a podcast-style recording. Listeners notice when Chapter 5 sounds noticeably brighter or quieter than Chapter 4, even if they cannot name why.

A concrete template example: your Audacity project template should include the project rate set to 48,000 Hz, the track set to mono, and your effects chain saved as a preset. When you open the template, record your chapter, apply the chain with one click, and export with your saved export settings. The entire process should take under two minutes per chapter after recording. If it takes longer, your template isn’t complete—go back and save the missing pieces.

Final Thoughts

The best settings are the ones you stop thinking about. Once your noise floor is clean, your levels are consistent, and your EQ is neutral, the technology disappears—and the story takes over. That is the goal for your listener, and it should be the goal for your settings.

Start with Audacity’s defaults, apply the four-step chain (noise reduction, normalize, light compression, loudness normalize), and test at 1.5x speed. Then adjust based on what you hear. Your voice, your room, and your microphone are unique; the settings are just the starting point.

If you’re producing an audiobook and want to hear how professional narrators handle pacing, dynamics, and clarity, listening to a well-produced title is the best education. Pick a book narrated by someone like Julia Whelan or Steven Pacey and study how their recordings sound at normal speed and at 1.5x. That listening practice will inform your settings more than any preset chain ever will.

Listen to this on Audible — Start your free trial and get two free audiobooks.