Feature: Improved Handling of Massive Numbers of Transcription Files in Videogrep

[neopiccolorat]

In the current version, Videogrep requires a video or audio file as input, which must also have the same name as the accompanying .srt, .vtt, or .json transcription file. While this setup is logical when working directly with media files, it creates unnecessary dependencies in certain workflows.

The video and audio file has to be there if someone is exporting a supercut or all the individual segments.

For example, when generating an XML file, the source video isn't really needed imo, only the transcription file is needed, since they have the same name. Therefore, I propose adding an option for Videogrep to work directly with transcription files, bypassing the need for the video or audio file altogether.

This feature would be extremely beneficial, especially since exporting XML files is a very fast process.

Additionally, for users who have transcription files stored in directories separate from their video files, batch processing an entire folder of transcription files would greatly enhance usability and efficiency.