Faster-Whiper-XXL Pro - Subtitle Edit integration?

[carusat]

Is Faster-Whisper-XXL Pro automatically integrated with Subtitle Edit 4.0.12 (installed on Win 11 Pro 64-bit, Nvidia 1080 Ti graphics) or would I have to use command line for every video file to get the speech to text generated subtitle? I'll happily donate but I'm a little scared of the learning curve. My use case is that I have a very large collection of documentaries, mostly English audio, but maybe 10% English + German or English + Spanish (viewed using Plex on AppleTV 4K, but I also have the above Win 11 Pro PC connected to my receiver/tv so I can switch to the PC to play the files directly using a player that fully supports primary and secondary subtitles (PowerDVD) when absolutely necessary). Many of these video files (almost all .mkv) do not have subtitles, so I generate 50 subtitle files or more per month. I would love it if I could output one .srt subtitle file in English (English transcribed, second language translated/transcribed to English). In the worst case I can use MKVToolnix to add a second text track to the mkv container, but Plex doesn't always handle forced tracks+non-forced tracks well so then I have to use a different video player. Either way, an even faster, even more accurate Whisper XXL is very attractive to me. I love your work by the way. Finding Whisper was a virtual Godsend! About 3 minutes to create a subtitle from 1 hour video length is really great. Can you help a mildly technically literate guy with this use case?

[Purfview]

Is Faster-Whisper-XXL Pro automatically integrated with Subtitle Edit 4.0.12 (installed on Win 11 Pro 64-bit, Nvidia 1080 Ti graphics) or would I have to use command line...

It has same file structure as regular version, you only would need to copy it to its Subtitle Edit folder.
So, the "Pro" usage is exactly same.

IMO, better to use it directly, as with "One Click Transcribe.bat" it's much simpler and SE can have inconsistent behaviour in some cases.

Edit the commands in the bat file, the line below ":: Start processing":

Replace: -m medium with better model -m large-v2
Add better VAD: --vad_method pyannote_v3

Replace: --standard with --sentence --max_line_width 46 --max_comma_cent 60
[Note to above: Because currently --realign has bug if a sub has break AKA more than a one line]

Add Pro features: --ff_vocal_extract mb-roformer --roformer_vram 8 --realign or --ff_vocal_extract mdx2_kim2 --realign
[Note to above: Maybe look at this issue: https://github.com//discussions/472 ]

If you do translation then add: --task translate

Add language of your file: -l German

You can remove -f json srt if you want only "srt" file.