Every chapter in context
Import the whole book, preserve filename order, assign opening, chapter, closing and sample roles, then see project-level readiness instead of isolated file scores.
Batch audiobook workflow · local QC · non-destructive prep
Audiobook Production & ACX QC Studio
Check every chapter together, listen where the measurements point, prepare reversible fixes and export a delivery package—without uploading your recordings.
Whole-book projectAudio stays localBefore / Prepared / DifferenceMP3 + reports
WHOLE-BOOK PREFLIGHTLOCAL
RMSReadyReadyFixReadyPeakReadyFixReadyReadyRoom toneReadyReadyListenReady
One decision surface for ordered chapters, technical targets, session consistency and the human delivery checklist.
MEASUREMENT MAP
Whole-book Technical Preflight
Turn a signal into evidence you can inspect. A useful analyzer connects the raw trace, the measurement window and a clearly labeled result without pretending that one number explains the whole recording.
Unordered chapter recordingsMeasure, compare and prepareOrdered delivery package
Open the full signal breakdownBest source, processing logic, limits and three-pass listening check
- Unordered chapter recordings
Start with the right material
Independent narrators, authors and producers preparing opening credits, chapters, closing credits and retail samples for ACX-style technical submission requirements.
- Measure, compare and prepare
Understand the transformation
Files are decoded sequentially, measured for RMS, sample peak, estimated quiet-floor behavior, edge room tone and cross-chapter consistency, then optionally prepared with bounded edge spacing, linked dynamics, target gain and peak protection.
- Ordered delivery package
Verify the usable result
Before, Prepared and Difference audition for the selected chapter plus ordered 44.1 kHz CBR MP3 files, optional WAV review masters, M3U8, JSON, CSV and an HTML QC report.
Human listening and official review remain
Where this workflow stops
Noise floor is an estimate rather than an automatic pass/fail authority. The browser cannot verify pronunciation, spoken credits, rights, editing quality or marketplace acceptance, and the tool is not affiliated with or endorsed by Audible or ACX.
THREE-PASS CHECK
Make the final decision by ear
- Use the same window and settings when comparing two files.
- Treat estimates as evidence, not as a substitute for listening.
- Export or note the measurement context alongside the result.
Quick answer: AudioWrench checks chapter loudness, sample peaks, noise-floor estimate, opening and closing room tone, duration, channel consistency and delivery format in one local project. It helps prepare files; it does not replace official review or human listening.
What the audiobook QC studio checks
Auditable technical measurements
Read decoded full-file RMS, sample peak, clipping candidates, a clearly labelled noise-floor estimate, edge room tone, duration, channels and sample rate.
Cross-chapter continuity
Flag level, room-tone and voice-color outliers that can reveal a changed microphone, distance, room or recording session even when each file passes alone.
Preview before export
Safe edge, gain, linked compression, channel-layout and resample operations are visible before processing. Noise and edit defects stay listening decisions.
Before, Prepared and Difference
Audition a selected chapter and inspect the metric change. A level-matched Before option avoids mistaking a louder result for a better one.
Open delivery package
Export ordered 44.1 kHz CBR MP3 chapters, optional WAV review masters, M3U8 listening order and JSON, CSV and printable HTML reports.
Current ACX submission requirements
| File organization | One chapter or section per file; opening and closing credits are separate files. Delivery filenames use standard US alphanumeric characters and identify the section. |
|---|---|
| Duration | No longer than 120 minutes per file. |
| Channel layout | All submitted files are consistently mono or consistently stereo. |
| RMS | Each file measures between −23 dB and −18 dB RMS. |
| Sample peak | Peaks do not exceed −3 dB. |
| Noise floor | No higher than −60 dB RMS. AudioWrench labels its automatic value as an estimate and asks for a representative room-tone listen. |
| Beginning and end | 1 to 5 seconds of clean room tone at the start and end of every file. |
| Delivery encoding | 44.1 kHz MP3 at 192 kbps or higher, constant bit rate. |
Before delivery, compare your files with the current ACX audio submission requirements and the ACX Audio Lab FAQ. Platform requirements can change.
How to prepare audiobook chapters
- Add the whole book. Choose finished opening credits, chapters, closing credits and a retail sample. Files stay on the device and are decoded sequentially.
- Review what matters. Filter Ready, Safe fix and Listen manually states. Correct titles, roles, inclusion and order before processing.
- Listen at the chapter level. Use the waveform overview, issue explanations and Before / Prepared / Difference player to validate every proposed change.
- Check consistency and content. Compare chapters against the book median, then confirm credits, section structure, spoken headers, sample and rights manually.
- Prepare the delivery package. Render chapters one at a time, verify the chosen layout and encoding, and keep the JSON, CSV, M3U8 and HTML audit trail.
How RMS, peak, noise floor and room tone differ
RMS describes average signal energy
The published target concerns each complete file. A chapter can have acceptable peaks and still be too quiet or too loud on average.
Sample peak catches the tallest decoded sample
It is not an oversampled true-peak claim. Peak protection controls headroom while the RMS plan deals with average level.
Noise floor needs representative silence
The automatic estimate uses quiet decoded windows. Air conditioning, computer fans, distant traffic and microphone self-noise still require listening.
Room tone is duration and continuity
The studio measures consecutive quiet windows at both edges. Room tone should sound like the recording space, not digital silence or a repeated artifact.
Consistency is a whole-book property
Median and spread views reveal chapters that diverge from neighboring sessions even when their individual technical values are inside a broad target.
Encoding is checked at delivery
Input can be browser-decodable audio. Final chapter files are rendered to the chosen CBR MP3 bitrate and 44.1 kHz sample rate in the export phase.
What this tool cannot verify
Performance and pronunciation
A waveform cannot know whether narration is compelling, a name is pronounced correctly or emphasis matches the manuscript.
All edit defects
Clicks, breaths, repeated phrases, missing words, plosives, page turns and room changes can escape automated measurement. Listen through every file.
Rights and official acceptance
The content checklist records your confirmation; it does not determine ownership, authorization, marketplace policy or an official review outcome.
Audiobook Production & ACX QC Studio: quick answer and technical limits
Quick answer: A local whole-book preflight and preparation workspace that orders audiobook chapters, measures delivery constraints, explains chapter-to-chapter consistency and builds an open delivery package without uploading narration.
- Best for
- Independent narrators, authors and producers preparing opening credits, chapters, closing credits and retail samples for ACX-style technical submission requirements.
- How it works
- Files are decoded sequentially, measured for RMS, sample peak, estimated quiet-floor behavior, edge room tone and cross-chapter consistency, then optionally prepared with bounded edge spacing, linked dynamics, target gain and peak protection.
- What you get
- Before, Prepared and Difference audition for the selected chapter plus ordered 44.1 kHz CBR MP3 files, optional WAV review masters, M3U8, JSON, CSV and an HTML QC report.
Know before you use it: Noise floor is an estimate rather than an automatic pass/fail authority. The browser cannot verify pronunciation, spoken credits, rights, editing quality or marketplace acceptance, and the tool is not affiliated with or endorsed by Audible or ACX.
Privacy: Each selected chapter is decoded, analyzed, prepared and encoded in the current browser tab. Metadata-only project state can be saved locally; source PCM and narration are not uploaded to AudioWrench. Privacy details →
FAQ
Does this audiobook studio upload my narration?
No. Selected chapter audio is decoded, analyzed, auditioned and encoded locally in the browser. Normal page assets can still be requested.
Does a ready result guarantee ACX acceptance?
No. The studio checks published technical targets and exposes its measurements, but official review and human listening remain authoritative.
How accurate is the noise-floor result?
It is a low-percentile decoded-PCM RMS window estimate. Confirm it with representative room tone and headphones; it is not a certified meter reading.
Can it process a complete audiobook at once?
It analyzes files sequentially and prepares exports chapter by chapter. A device-aware package plan chooses ZIP or report and one-by-one fallback instead of one fixed upload-size limit.
What can this tool not verify?
It cannot verify pronunciation, performance, rights, spoken headers, missing text, every edit defect or official acceptance. These remain human checks.
What does the delivery package contain?
The package can contain ordered 44.1 kHz CBR MP3 chapters, optional WAV review masters, M3U8 listening order and JSON, CSV and HTML QC reports.
Focused tools for chapters that need repair
Keep the whole-book studio as the project map, then route only the affected chapter to a focused local repair tool.