Booth Ready · Proofing
Hear every misread before ACX does.
Send your recorded audio and the script you read from. We return your audio with audible beeps at every divergence — every dropped word, every misread name, every mispronunciation — plus a prioritized written report. You hear the problem and know where to fix it, in 90 seconds of listening.
What the marked audio sounds like.
A real proof run on Chapter 4 of Beowulf: A Modern Retelling, narrated by Mike Vendetti. The first 90 seconds of the take — long enough to hear three misreads and one pronunciation flag.
1. Original recording
The take as recorded. Sounds clean. Listen first.
2. Marked audio (what you receive)
Same 90 seconds with 880 Hz beeps at every divergence the proofer found. Five beeps in this clip.
What the beeps caught in this clip:
- 00:07 — misread: script says “wary silence,” narrator carried over the chapter title and read “wary challenge.”
- 00:14 — misread: script says “carried in platters,” narrator read “carried the platters.”
- 00:44 — misread: script says “the smell of good food,” narrator read “a smell of good food.”
- 01:13 — phrase variance, recommended for review.
- 01:37 — pronunciation flag: Wealhtheow read as Waldtréo (narrator choice; flagged for series consistency, not retake).
A typical reader misses two of those on a single listen. The proofer catches them in 90 seconds, with timestamps. The full report covers the entire 15:48 chapter — 35 divergences total, tiered by severity.
Want to see the engine at full scale?
Six chapters of Beowulf, 1:53:52 of audio, 15,726 words, 222 divergences flagged with timestamps and per-chapter PDFs.
See the 6-chapter Beowulf demo →Three steps.
Same workflow whether it's an audition take or a 200,000-word manuscript. The engine doesn't care about length — only the file count changes.
Send audio + script
Upload your WAV or MP3 — any format that's playable. Send your script in DOCX, PDF, RTF, or TXT. We accept ACX submission spec but anything mono or stereo, 22 kHz or higher, will proof cleanly.
Forensic diff runs
Word-level transcription, compound-word-aware comparison against the script, pronunciation watchlist for proper nouns and foreign vocab. Every divergence gets a timestamp.
Receive marked audio, DAW markers, and report
Your audio comes back in the same format you submitted (WAV in, WAV out — MP3 in, MP3 out) with 880 Hz beeps at every pickup point, plus a .cue sidecar that loads into Reaper, Audition, Pro Tools, or Studio One as named markers — jump pickup-to-pickup with one keystroke. Plus a written report sorted into must-fix, recommended, and observation tiers.
Already discussed with Mike?
Pay now via PayPal.
Skip the intake form if you and Mike have already aligned on scope. These buttons drop you straight into PayPal at the agreed amount.
Audition Proof — $50
Flat rate. ≤ 3,000 words, ≤ 20 min audio. Same-day on weekdays.
Chapter or Full-Book Proof
$8.00 / 1,000 words × complexity tier. Set your word count and SKU in the calculator above, then pay your quoted total here.
Pick the Chapter or Full-Book SKU in the calculator above.
Reproof (after pickups)
50% of original proof price, floor $25. Re-run after you record corrections to confirm each pickup landed.
Pick the Reproof SKU in the calculator above.
Not sure what scope fits? Use the intake form and Mike will confirm the quote before any charge.
Common questions
Will my audio sound different after marking?
No. The marked file is your original audio with short 880 Hz beeps inserted at pickup points — the recording itself is untouched. The beep is loud enough to hear clearly through monitoring headphones but never overlaps the spoken word. You delete the marked file once you've recorded your pickups; the master stays clean.
What format do I get the audio back in?
The same format you sent. WAV in → WAV out (same sample rate, same channel count). MP3 in → MP3 out (re-encoded at the same bitrate, mono if you sent mono). If you want a different output format than what you submitted, note it in the comments and we'll match.
What if I recorded multiple takes of a scene?
Send the take you want proofed — usually your tightest. If you're not sure which take to commit to and want a sanity check on two of them, send both and we'll proof the cheaper one as an audition ($50) and the other as a full proof. Often the takeaway is "your second pass was cleaner than you think."
Does this replace having a human proof-listener?
No. It catches what an automated diff can catch — dropped words, misreads, mispronunciations against the script, watchlist hits on proper nouns. It does not catch performance choices a human would flag: pacing, emotional register, mouth noise tolerance, whether a line landed. Use this to find the mechanical errors so the human can focus on what only a human can hear.
What if my script has a typo?
The diff will flag it — your audio "diverges" from the script even though your read is correct. The report tiers these as observations (lowest priority) so they're easy to dismiss. If the manuscript has dozens of typos, you may want Booth Ready prep first; it's harder to proof against a sloppy script.
Can I get just the report and skip the marked audio?
You can, but it's not cheaper. The pipeline produces both as one pass — the report references timestamps that match the beeps. Most narrators say the beeps are the part they actually use; the report is for archive and for talking to your editor.
What about ACX rejections — can I get a proof after I've been rejected?
Yes, and this is exactly when proofing earns out. Submit the rejected file with the script and ACX's rejection notes if you have them. The report will tell you whether the issues are mechanical (proofing fixes) or technical (mastering, denoise, EQ — proofing doesn't touch those, but we'll flag what needs an editor's hands).