This week: New Service: Advanced Audiobook Proof Check See Details

Audiobook Proofing: Why It Takes Days, and How It Became Minutes

Proofing is the least glamorous stage of audiobook production and one of the most expensive in hours: someone sits with the manuscript and listens to the entire book in real time, marking every place the narration and the text disagree. An 18-hour book is, at minimum, 18 hours of concentration — usually more, because you stop, scrub back and note timestamps. Professional proof-listening is commonly priced around £12 per finished hour, so a 10-hour book runs in the region of £120 and takes days to schedule and complete.

What proofing actually catches

Strip away the mystique and nearly every proofing mark is one of three things:

  • Substitutions — the narrator read a different word than the manuscript (“mantel” for “mantelpiece”, “car” for “carriage”).
  • Omissions — a word, phrase or whole line skipped.
  • Insertions — a word added that isn’t in the text.

Each one needs a timestamp, the manuscript line, and a decision: is this a re-record, or is it fine? Then every accepted re-record has to be compiled into a pickup list for the narrator — chapter, time, reason, and the exact sentence to read again.

Problem solved

Our Advanced Audiobook Proofing tool transcribes your entire book  — your files are never handed to a third-party cloud service — and compares every word against your manuscript. The comparison is built for narration, not for text: numbers, roman numerals and pronunciation variants are matched by meaning, so “forty” read for “40” is never flagged. It was tuned on real audiobooks with a simple goal: flag what a human proofer would mark, and nothing else.

The part a machine should never do

The judgement. Every catch is presented as a card — the manuscript line against what was narrated, with the timestamp and a confidence score. You play the moment and decide: accept for re-record, or dismiss. Nothing goes on the pickup list on a machine’s say-so.

What you walk away with

Once you’ve reviewed the catches, the deliverables build themselves: a numbered pickup script in book order (chapter, time, reason, the full sentence for your narrator to re-read), a preview audio file with the original take of each pickup so the narrator can match tone and pace, and a downloadable report with timestamp and chapter in separate columns, ordered by book position, ready to edit from. A manual proofing booth is included alongside, for the marks only ears can make.

The economics

Advanced proofing is priced at £19 per audio-hour (minimum £39) — a 10-hour book is £190. Manual proof-listening at around £120 buys a single pass through the same book over several days; this is the whole book in minutes, with the pickup script, audio references and edit-ready report generated as part of the job, and the manual booth included alongside. The first 20 minutes of any book are proofed free, so you can watch it catch things on your own narration before paying anything.

Audiobook Loudness: Why Platforms Reject Quiet Files (and the Fix)

Loudness is the most common purely technical reason a finished audiobook gets bounced — and the most misunderstood. Your narration can be beautifully recorded, cleanly edited and still rejected, because distribution platforms don’t judge sound with ears: they measure it, file by file, against a delivery specification.

The spec, in plain English

The standard delivery loudness for audiobooks is −18 LUFS integrated with a true-peak ceiling of −3 dB. LUFS is a measure of perceived loudness over the whole file; true peak is the highest instantaneous level once the audio is reconstructed by a player. Together they say: the book should sit at a consistent, comfortable listening level, with enough headroom that nothing clips downstream. This sits inside ACX’s required range for typical narration, and it’s what publishers and platforms commonly specify on delivery sheets.

How books drift off spec

Almost never in one dramatic way. Chapters get exported across weeks of sessions at slightly different levels. A re-recorded chapter comes back hotter than the original ten. A whole book is simply mastered quiet — −24 LUFS narration is very common from home studios. Each file sounds fine on its own; measured against the spec, half the book fails, and listeners notice the seams between chapters even when reviewers don’t.

What correction looks like

Level correction is a solved, verifiable problem — as long as it stays level correction. Our Audiobook Loudness Correction tool does it for you on the spot, so you don’t need to wait. it measures every file, brings anything off-target to −18 LUFS with peaks under −3 dB, then re-measures its own output so you download verified numbers, not hope. Files already on target are returned byte-for-byte unchanged — no pointless re-encode, no quality tax on chapters that were already right.

The caveat nobody prints: noise comes up too

Turning a quiet recording up turns its room noise up with it. If your noise floor is borderline, correcting the loudness can push the hiss past a platform’s limit — that’s physics, not a software failing, and any tool that claims to fix loudness and noise in one automatic pass deserves suspicion. The honest workflow: measure first. Our free compliance report shows the noise floor of every file in your book before you spend anything, so you know whether you need level correction, or a quieter recording.

The economics

Correction is priced as the utility it is: £9 per audio-hour, minimum £15 — a 10-hour book is £90 — and the first 20 minutes of any book are corrected free so you can verify the result on your own narration first.

Now Live: the ST Online Audio to MP3 Converter

ST Mastering has launched an Audio to MP3 Converter service that runs 24 hours 365 days a year online service that processes your files instantly — studio-grade conversion of WAV, AIFF and MP3 files into clean, distribution-ready MP3s in seconds in your browser. No apps or extensions to install, just a simple drag and drop service.

Choose 320, 192 or 128 kbps, drop in one file or a whole batch, and download the results immediately. Your first three conversions are free with an account, and passes cover unlimited use from £6.99.

It exists because still today, if we need to convert a file to mp3, we still need to choose an application, set the application up, choose the spec, save the file somewhere, find the file etc. And usually we are doing this from a desktop or laptop computer. It’s absolutely still a chore of a task to complete for a simple file converting process.

We’ve made that a bit easier, with a web service you can use from anywhere and from any device. Super handy if you are needing to convert something from files you just received on your phone. No need for applications, laptops etc. Just have your file ready, and our service does the rest.

Sharing MP3s shouldn’t stall because you havent got around to converting your audio to mp3, so you get platform-ready MP3s out, no installs, no ads, no quality issues.

Try the Audio to MP3 Converter →