The brief
A language service provider needed a multimedia partner. Their end client runs a library of internal informational courses, the LSP handles the translation, and someone has to turn those translations into finished audio.
The constraint was the voice. The courses are updated continually, and every update has to sound like it came from the same person as the last one — not for polish, but because a consistent voice is what lets an update be a patch rather than a re-recording. Change the voice and the whole course has to be redone.
What made it difficult
Casting for the long term, not for one job
We serve as an extension of the LSP, handling the audio and multimedia implementation for the translations they deliver. Casting draws on a deep talent roster with recurring availability in mind: the question is not only who is right for this recording, but who will still be there for the next one.
Recording only what changed
When a course is updated, we record the updated segments rather than re-recording the entire course or video. That is only possible because the voice, the direction and the delivery match the original session — the new segments have to drop into the existing audio without a seam.
Keeping the schedule when the talent cannot
Talents are occasionally unavailable. Where the project allows it, pre-approved alternative voices keep the work moving instead of stalling it, so an urgent update does not wait on one person's calendar.
Quality control
As on every JBI project, the audio was checked by a native-speaking QA reviewer against the script: pronunciation, including the project's own pronunciation guidelines, plus tone, style and accuracy. Deliverables are mp3.
The result
The same voiceover talent, held across more than four years of recurring content. Production overhead came down sharply, because updates re-record segments instead of courses. And urgent update schedules were met — including over holiday periods, when talent availability is at its thinnest.
