Format
Timestamps make a transcript navigable
A transcript without timing is text you can search but not use. Put a timecode on every segment and it becomes something you can jump to, quote, and check.
A wall of text is not a transcript
Strip the timing out of a transcript and what is left is a block of text - searchable, maybe, but not navigable. Finding the ninety seconds that matter means reading the whole thing again, or scrubbing the recording by hand and guessing.
That is the gap between having the words and being able to use them. The words alone tell you what was said. They do not tell you where.
Per-segment timing points at a moment
Vocateca keeps a timestamp on every segment, not just a start time for the whole file. Click a line and the player jumps to that moment - you go from reading a claim to hearing it said in under a second.
That is what makes a transcript navigable rather than just legible. You are not reading a copy of the audio; you are using an index into it.
A citation needs a timecode
Quote a sentence from an hour-long recording without saying when it was said and the quote cannot be checked - the listener has to take your word for it, or scrub the whole file to find it themselves.
Add the timecode and the quote becomes verifiable. Anyone can jump to that second and confirm the words are actually there, in that order, said by whoever you say said them.
SRT and VTT carry the timing forward
Export a transcript from Vocateca as SRT or VTT and each line keeps its start and end time - the format a caption track or a subtitle file needs to sync with the recording. Export as JSON and the same segment-level timing survives for anything you build on top of it.
Timestamps are not a display feature that stops at the app window. They are data that travels with the transcript into whatever you export it for.