I’ve had a pipeline in mind for exactly this purpose that I want to build when I get around to it:

• Download the audio file from RSS feed
• Self hosted AI transcription model (with output that includes timestamps)
• Self hosted LLM to recognise ad sections and return the start and end timestamps as json
• ffmpeg to slice those timestamps out and stitch the rest back together

In theory, this should be able to remove ad and sponsor sections of any length completely automatically and there’s nothing to stop it working on videos too

Every user has to self host their own?

Did… Did you see what community you’re in?

I’ve found Android TV to be the most usable TV OS tbh. I use Konstakang’s LineageOS Android TV 15 image on a Pi 5 which is source available (non commercial only licence). And Projectivy Launcher (closed source but is by an indie dev and better than the stock Google one). The Pi is CEC compatible so I can control it with my TV remote no problem, and I use Moonlight to stream games from my PC.

If you already have hardware, there may be an Android x86 TV release somewhere but I haven’t personally tested any, and you have to make sure the apps you want support x86 (all the open source ones like Jellyfin should)

Edit to add: I also personally haven’t found a need to install GApps as all the apps I use are either open source, or were made to work also on FireTV so don’t rely on GApps APIs. (Use SmartTube instead of YouTube, it’s a better experience anyway)

I built a near identical server for my parents and just sync my nextcloud folder to theirs using syncthing