noScribe Editor: Edited speaker labels aren't exported to a .vtt file.

[BabyFnord]

Describe the bug
Edit: This is about assigning proper names to voice-tags. Replacing the generic S0n speaker labels with real speaker names in noScribe Editor, unfortunately, doesn't export to the .vtt … This is a problem with subsequent analysis.

Could this be fixed, pretty please? Thank you

Screenshots

Video Filename

NOTE
Transkribiert mit noScribe Vers. 0.6
Audiodatei: /video.mp4
(Start (hh:mm:ss):  00:00:00 | Modell: precise | Sprache: Auto (auto) | Sprecher:in erkennen: auto | Überlappende Sprache: 1 | Zeitmarken: 1 | Füllworte: 1 | Pausen markieren: 1)


NOTE media: /video.mp4

1
00:00:00.000 --> 00:01:16.400
<v >(1 Minute[n] Pause) 

2
00:01:16.400 --> 00:01:28.180
<v S04>Sprecher: 

3
00:01:16.400 --> 00:01:28.180
<v S04>[00:01:16]

4
00:01:16.400 --> 00:01:28.180
<v S04>Text …

System (please complete the following information):

• OS: macOS Sonoma 14.7.6
• noScribe Version: 0.6

[kaixxx]

Hi @BabyFnord
Yes, you are right, the underlying speaker tags are not updated. I have to think about how to implement this in the UI so that the user can still control what is happening.

[BabyFnord]

On the same note, could it be made possible to edit start times in noScribe Editor, too? The case being documentaries in which music is sometimes played in between, here noScribe seems to miss the entry time of speech by several seconds. My attempts to edit speaker labels and edit timings with other tools, like converting between .vtt and .ass and back, ended up in messy files.