Voices kept as series assets
Each character’s voice is assigned once per language and reused, so episode sixty sounds exactly like episode one.
For animation studios
Animation travels further than almost any format: characters are drawn rather than cast, and audiences readily accept a dubbed voice as the voice. Yet most series reach only the markets that could fund a studio cast. Octavia voices each character consistently across a run, keeps the score and effects, and leaves the songs to people.
The specific reasons this content usually stays in one language.
For a preschool audience a subtitled version is effectively no version — no dub means the show isn’t in that market.
Seventy-eight episodes recorded across years and studios can leave one character sounding like three different people.
Animation is cut to the source dialogue, so a translated line that runs long has no picture to stretch into.
The parts of the platform that matter for this work, rather than the full feature list.
Each character’s voice is assigned once per language and reused, so episode sixty sounds exactly like episode one.
Speech isn’t sped up to fit. An adapter tightens the translated line at review instead, so it lands on the cut without sounding rushed.
An in-market reviewer checks vocabulary, register and anything culturally off for the age group before a single line is voiced.
Character names, catchphrases and invented words pinned one approved way per market — matching the packaging, books and app.
Dialogue goes over your M&E; for older seasons that survive only as a mix, Magic Mode separates voices from score and effects.
Whole seasons and short-form spin-offs processed together, so a new market launches with enough episodes to schedule.
Four steps from the file you already have to every language you need.
Dialogue, music and effects stems where you have them; separation covers seasons that only exist as a mix.
A voice for every recurring character per language, with names, places and catchphrases pinned.
An in-market reviewer adjusts lines for length, age and culture before any audio is generated.
Dubbed episodes and subtitle tracks per market, with the same cast from first episode to last.
One upload in, every format you need to publish out.
Including the ones where the honest answer is a limitation.
Leave them to people. A theme tune or an in-episode song needs a lyric adaptation that rhymes, scans and fits the melody, then a singer — a separate creative job, and not one Octavia does. The workable pattern is to dub the dialogue and drop in songs your music team has adapted and recorded per language, or subtitle the original song where the format allows.
Lip-sync is in beta and we would not schedule a delivery around it. Animation is more forgiving than live action here: mouth shapes are simplified, and audiences of dubbed animation accept loose sync. The realistic aim is lines that start and stop with the mouth, which comes from adapting line length at review rather than altering the picture.
Only with each performer’s written consent, and check their contracts and any union terms first — most voice agreements for existing series predate synthetic voices and do not cover them. Where a character is voiced by a child, consent runs through their guardians and deserves extra care. Many studios assign a professional synthetic voice per language instead, so the character sounds native in each market.
In your largest markets, probably not on its own. Where the dub is a performance in its own right — comedy built on timing, a celebrated lead, a theatrical release — a studio cast and voice director remain the right call. The fit is everything else: the long tail of markets, older seasons, shorts, spin-offs and educational series, where a cast per language never made financial sense.
Practical guides from the Octavia editorial team.
The same engine, pointed at a different kind of work.
Start with a short-form spin-off in one new market — the lowest-risk way to hear your characters in another language.