• 1 Post
  • 42 Comments
Joined 1 year ago
cake
Cake day: August 7th, 2024

help-circle
  • Thanks for the recommendation, it certainly seems like an interesting project, although it’s current capabilities are almost backwards from what I actually want. My current workflow is:

    • Listen to my library in Navidrome via Feishin/Tempus
    • Scrobble my listens to ListenBrainz
    • ListenBrainz generates my weekly recommendations playlist (things not in my library)
    • I listen to the playlist on the ListenBrainz site via YouTube embeds
    • Any songs I like, I download high quality FLAC files of using slskd by buying legitimately

    Just being able to see my recommended songs in Tempus would remove some of the barrier of having to log in to ListenBrainz every week (which I often push to the bottom of my to-do list and end up missing recommendations). I don’t even really need to stream them directly in the app, just being able to see them and open the YouTube link would be a good start.

    I get that this might be a bit of a niche way of doing things and everyone will have a slightly different idea though, I don’t really expect my exact personal workflow to be catered for by open source devs.












  • Data centers use massive amounts of power. Reducing that power consumption would be a net positive on the environment.

    That said, looking at the site, I can’t see anything that suggests they’ve done any research into that. Instead they seem to be promoting their research into keeping kids safe online, with the only mention of the OSA being a short description in their recent June report with no comment about anything to do with it’s obvious shortcomings.







  • It sounds to me like one (or more) of your containers is referencing something on your storage drive, but Docker is loading before your drive gets mounted. When Docker sees that the folder its trying to access doesn’t exist, it creates it, blocking your drive from taking that name.

    To fix it, you would need to make sure your storage drive is mounted before Docker starts, how you do that is down to you and your particular setup though.




  • 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