iPolloWork Video Voiceover
Use for requested new/revised speech or its captions. Existing audio, enabled voice settings and music-only work do not request synthesis. Read shared session boundaries once when unknown, affected Voiceover sections and exact current voiceover.json/storyboard/scene anchors. Respect authorization, saved disabled state and pinned/per-frame voices.
Synthesize through the built-in batch contract; mount actual returned audio/captions and apply measured timing shifts once while preserving prior speech until success and retaining music/SFX. Captions-only work reuses saved alignment. Do not guess provider/voice, fabricate timings or replace the TTS service. Save source for the client's combined gate; disclose unfinished required speech without creating a second check loop.