Is there a way to integrate Speaker Diarization / Speaker Identity to the generated ASR subtitle?

[Xeynt]

The purpose of this is so that we can prompt the 3rd party translation ai model to take into consideration the speaker id when translating the subtitles. Very often when the subtitles are translated from another language to english, the "I", "You", "We", "It" gets swapped around. It's as though the translating LLM multi context model isn't aware who's speaking so it's simply doing a guessing work.

So basically when translating, have Faster-Whisper include the speaker id, it could be "Speaker 1" etc.
Then, in the prompt fill in LL Player's Translation tab, include a prompt to ask the translation model to take into considerations the speaker id.

I've searched abit and it seems like there are 3rd party add-ons for whisper that is able to get this working but as I'm still a little new to all this, I'd like to check if it's possible to modify the one that LLPlayer?