I research far-right groups who are mostly banned from YT. I want some kind of tool that automatically downloads content from their groups (mostly on BitChute, Rumble, Oddysee) and then does a TTS transcribe if subtitles are unavailable.
I’ve tried Metube, which is really good at downloading from niche sources but lacks the concept of a subscription. In other words, I cannot just give it a bunch of channels I want and then let it do its thing.
Yt-archivist is excellent but is YouTube only. It doesn’t support the niche video sites I need.
If only there were a tool with archivist’s features but Metube’s breadth of support. I’ve not yet seen any tool that can automatically transcribe the content in a directory, but that would be perfect - some tool that I could feed a directory of content and then have it automatically transcribe anything that doesn’t have subtitles.
The reason I need transcription is so that I can search for certain comments. For example, I might want to find out when a particular group started campaigning on an issue or used a particular phrase.
Ideally, I’m looking for stuff that runs as a Docker-Compose configuration. I have a NAS with a few TB of storage. It would be great if I can just run this directly on the NAS and
my tool (https://prometheus.tube) does this. you feed it URLs, and it grabs the videos periodically.
Is this tool Docker installable? Is there a GitHub link?
The link you provided above appears to be a running instance of some very impressive software. Thank you for sharing.
Is transcription something you would consider adding later? Ideally, this would be something where I can provide AWS or Google Cloud credentials, and it would offload the transcription to a cloud engine.
Does your service allow subscriptions of alt-Tubes? (e.g. BitChite, Rumble)