The Master Guide / 10 Voice and lip sync · Voice and Arabic dialects
Three texts per line
Every line exists in up to three forms. Keep them apart, because each goes to a different place.
Every line exists in up to three forms. Keep them apart, because each goes to a different place.
Text | What it is | Where it goes |
The canonical line | The approved wording in the dialect's normal spelling, with no vowel marks and no tags: the client's script, inside «…». It never changes | Captions, client pages, the treatment, any video prompt that quotes a line |
The engine text | The canonical line with vowel marks, numbers written as words, English words in Arabic script, and English delivery tags | ElevenLabs only. Kept in your line ledger (the list of every line in your shot plan) with the engine, voice and date |
The Franco line | The approved take spelled in Latin letters, only for a shot where you chose Route 1's Franco quote | That shot's video prompt only; never a voice engine (see Speaking shots and lip sync → Route 1, for close-up hero lines) |
Captions and transcripts are always compared with the canonical line, never the engine text.