Cool! As soon as the plant identity database is selfhostavle I'll be digging into this!
Kuvwert
This is desperately needed.
I saw sponsorblock-ml and am playing with it using whisperx for transcript/timestamps and ffmpeg for cutting out the timestamps that were detected by sponsorblock-ml then reserving that audio as an rss feed.
It's not great so far though
I've been looking into sponsorblock-ml for an alternative approach
I'm running 3.1 8b as we speak via ollama totally offline and gave info to nobody.
Well, perceived damage anyway. I can't speak to how IP owners have been effected by LLMs, and I don't believe it would be easy to quantify.
From what I understand the next rounds of ai are being trained on further refined versions of the same datasets and supplemented with synthetic data.
The damage to existing copyrighted content is already done.
Source: I'm a random internet user
No shade on you, dude... but if it's not available in a public repo where people with more experience than me have the opportunity to validate and review it... then I'm really really really not interested in downloading or running it on my machine.
I have trust issues with cloudflare yes, but I also have trust issues with random zip files from strangers' cloud drives.
I appreciate your helpful attitude anyway 🙂
Does your tool have a public repo I can take a look at?
Tailscale is so cool too. I'll definitely be switching if I can ever use my own domains
What benefits are you having from switching?
I understand that this comment isn't helpful, but you're gonna save yourself a lot of headaches by just transitioning to DaVinci Resolve.