Serialize a compiled timeline's caption track as the one canonical WebVTT
file its caption deliverable must equal byte for byte.
The header carries the timeline id. Cues are ordered by start frame, then end
frame, then id in code-unit order, and each cue's frame interval maps to
whole milliseconds with nearest rounding on the exact production frame rate.
An authored top or bottom placement selects the first or last snapped line
of a horizontal cue. Omitted placement uses WebVTT automatic line placement.
The payload sits in a <lang> span, inside a <v> voice span when the cue
names a speaker, and every block ends with a blank line. Throws when an id is
not a WebVTT identifier, a language is not a well-formed RFC 5646 tag, the
timeline's fps disagrees with its exact frame rate, or a cue interval is
empty or collapses on the millisecond clock, or a placement is unsupported.
Serialize a compiled timeline's caption track as the one canonical WebVTT file its caption deliverable must equal byte for byte.
The header carries the timeline id. Cues are ordered by start frame, then end frame, then id in code-unit order, and each cue's frame interval maps to whole milliseconds with nearest rounding on the exact production frame rate. An authored top or bottom placement selects the first or last snapped line of a horizontal cue. Omitted placement uses WebVTT automatic line placement. The payload sits in a
<lang>span, inside a<v>voice span when the cue names a speaker, and every block ends with a blank line. Throws when an id is not a WebVTT identifier, a language is not a well-formed RFC 5646 tag, the timeline'sfpsdisagrees with its exact frame rate, or a cue interval is empty or collapses on the millisecond clock, or a placement is unsupported.