Take any stage, or the whole chain.

Hum, hiss, air conditioning, traffic rumble, mouth noise and plosives, out before any editing decisions get made. Done per file and matched across the session, so chapter fourteen doesn't sit differently to chapter two. Home-studio recordings tend to need the most work here.

Fluff-and-repeat or punch-and-roll, whichever way the narrator works. Retakes chosen and joined so the join is inaudible, breaths shaped, reduced or removed to the client's brief, page-turns and chair noise gone, room tone matched. Then levelled, de-essed and mastered as a whole title rather than chapter by chapter, so the sound holds consistent end to end, with a QC pass before it goes anywhere.

Listened against the manuscript, line by line, by a person who wasn't in the room for the record. Fresh ears catch what the studio stops hearing — in fiction especially, character names, accents and the nuances that have to stay consistent across a whole book. Misreads, dropped words, wandering pronunciation and takes that got left in.

Machine and human proofing run as two independent passes, and the results are combined into a single list. Software is relentless about dropped words, misreads and anything that deviates from the manuscript; the proofer judges intonation, character consistency and whether a flag actually matters. Neither pass sees the other's findings, so between them the catch rate is higher than either alone. Priced as an add-on to standard proofing — talk to us about whether a title warrants it.

(Machine proofing is available on its own on request, but we find it may only work well on non-fiction titles, or straightforward reads with common names — our experience has shown that errors can on occasion slip through.)

Retakes dropped in and rematched for level, tone and room, so the correction doesn't announce itself. From a proof list, a QC rejection, or an author who has had second thoughts about a paragraph. This is usually the stage standing between a title and its release date, so it goes to the front of the queue.

To all specs. Send the numbers you need to hit — retail platform, distributor, or your own house sheet — and files come back conformed, in your naming, ready to upload. Already have finished audio mastered to the wrong target? Send that instead and we'll remaster it rather than start the edit again.

Table of contents, chapter marks, opening and closing credits, retail metadata written to the files. Dull work, and the reason a lot of finished titles get bounced. Delivered by whatever route suits you — we fit into your pipeline rather than asking you to learn ours.

Solo & Dual Readers

Single narrator, or two voices alternating. On a dual read the work is keeping both even — level, tone, pacing — usually across two rooms and two recording days.

Multicast & Duet

Casts recorded separately, in different rooms and often on different days, assembled into scenes that sound like one session. Tone and level matched across every performer, overlaps timed so the dialogue plays.

Parallel Session Editing

Where it suits the schedule, we can follow the studio day — editing as the record goes rather than waiting for the wrap, so pickups get flagged while the narrator is still in the chair.

Restoration

Archive material, tape transfers and recordings that were never going to pass QC as they stand. Backlist titles brought up to a current spec without a re-record.

Multi-voice, FX & Music Placement

Full-cast and multi-voice spoken word: dialogue edited and balanced across performers, effects built and placed to suit the scene, and any music you supply fitted to the edit. Spoken word is the speciality — we'll happily take on sound post for screen or a video podcast, we just don't hold ourselves out as specialists in it. Composing and scoring we leave to composers.

Podcasts

Episode edits, FX and supplied music placed, loudness-compliant masters, at a turnaround that survives a weekly release.