ABSTRACT
01.What the original audience got
The first audience hears the actor's own voice, sees lips that match the words, and feels the timing the director chose: the pause before the punchline, the laugh that breaks the line, the music under the scene. Every other audience has usually received less: subtitles to read over the picture, or a stranger's voice with a mouth out of sync. The work was the same. The experience was not.
02.What the same means
| In a Familiar dub | |
|---|---|
| Voice | The actor's own voice, taken from the clip itself |
| Lips | Rendered to match the new language, so eyes and ears agree |
| Delivery | The same timing, emotion and emphasis as the original |
| Scene sound | The laughter, the music and bass, the sound effects, the ambience |
| Words | Names and catchphrases kept through a Do Not Translate list; jokes rewritten to land |
Lines that can land more than one way, a joke or a jab, come back as three translations with different intent and emotion. Creative directors pick one and edit any line against the original before the dub renders.
Each language arrives as a finished video, up to 4K, with the full mix as true 5.1 surround or stereo, in 24-bit PCM WAV. Subtitles come in .srt and .vtt from the reviewed transcript, with the translated title and description and the voice track alone for your own mix (the FAQ).
03.Thirty languages out, 99 in
Footage can come in from 99 source languages. Dubs come out in 30, any to any, in three quality tiers, best first and alphabetical inside each tier.
| Tier | Languages |
|---|---|
| Tier 1 | Cantonese, English, French, German, Indonesian, Italian, Japanese, Korean, Mandarin, Portuguese, Russian, Spanish, Thai, Vietnamese |
| Tier 2 | Bulgarian, Catalan, Croatian, Dutch, Greek, Lithuanian, Norwegian, Slovak, Swedish |
| Tier 3 | Arabic, Belarusian, Danish, Kazakh, Latvian, Serbian, Ukrainian |
The full table, with codes, lives in the languages docs.
04.Why most of the world never got a dub
Human dubbing is skilled work: translators, voice actors, directors and mixers for every language. That craft is why a good dub sounds right, and it is also why human dubbing was never usable for most of the world. Even top productions dub only a few languages, because of the cost, the people and the time. Smaller productions, shows and vertical dramas don't dub at all.
So the audiences a work could reach were set by its budget. A great performance in one language stayed in one language.
05.Machine learning for the tedious part.
The tedious part of dubbing is the volume: every line, in every language, voiced in the actor's own voice and synced to the mouth. Machine learning carries that. Nothing is created from a prompt and nothing new is made up; every frame and every sound answers to a line that was already performed. The choices that make a dub good, which joke lands, which tone fits, which name stays, remain with the people who made the work.
All of the data we trained with ourselves are actually generated by the Familiar team.
06.Every audience
When the same performance can reach every audience, the question changes from which languages a production can afford to which audiences it wants to meet. A vertical drama, a lecture or a feature can open in 30 languages with the actor still the actor.
We're another tool for the creative process.
QUESTIONS, ANSWERED
Does every language get the same quality?
All 30 output languages carry voice, lip-sync and the scene sound. They are grouped into three quality tiers, best first.
Is it still the actor's voice in every language?
Yes. The voice in each language is the actor's own, taken from the clip itself, with the same timing and delivery.
Who decides how a line is translated?
Your team. Lines that read more than one way come back as three options; creative directors pick one and can edit any line before the dub renders.
REFERENCES
Multimodal Translation
They're a team of 8 with PhDs, backed by Samsung, RedBird, Gold House. Our mission
FAMILIAR · THE LAUNCH FILM · 2:31