I installed the high-performance backend and the recording backend over the weekend. Both seem to work OK, although there are some lip sync problems with the short test recordings I’ve been taking. I’m running NC28.0.3.
I also have the Whisper transcription installed and running, and I’ve used it for some audio files in the past. For instance, from Zoom calls with recording, I get both a video file and an audio-only file. I have done transcriptions of the audio files.
Although I’ve been using NC for a couple years now, I’ve only used the recording and transcription a couple times, so I’d consider myself a novice in that regard.
The files that are recorded from Talk are full video files, and when I use the assistant to transcribe, it is only looking for audio files.
My question: How do I get a transcription from a Talk recording?
Ancillary question: Is the high-performance backend going to give me better performance in all situations? Even with participants outside my network? All my NC-related servers are on my home network.
Thanks in advance!