Available on plain rss https://anchor.fm/s/e8b6108c/podcast/rss too

Also, please send me suggestions for articles I should add there, because I have no idea how to approach curation of that library.

Any other feedback is appreciated.

Edit: I just found this https://lemmygrad.ml/post/6319. Looks excellent! I’ll work on getting it through.

Edit 2: and then I’ll probably come this way: https://dessalines.github.io/essays/dessalines_marxism_study_plan.html

Edit 3: I’ll also play with the order, because it feels like a good idea to mix introductions for multiple subjects, so that it feels a little less repetitive. Even though… can’t do much about the dryness of AI speech.

  • ReadFanon@lemmygrad.ml
    link
    fedilink
    arrow-up
    3
    ·
    1 year ago

    I’d recommend considering using Balabolka as a TTS generator because if you go to Tools > Use online TTS service you can select IBM Watson as the TTS engine (and it bypass the word limit restrictions.)

    I think that Michael is my preferred TTS option for Watson but it’s been a while. Whatever the case, Watson is far and away the most listenable TTS that I’ve come across.

    Lmk if you want screenshots because the Balabolka UI is a bit clunky.

    • nephs@lemmygrad.mlOP
      link
      fedilink
      arrow-up
      3
      ·
      1 year ago

      It may be a bit challenging because I don’t use Windows, and this project is kind of a useful early spinoff from few experiments I’ve been doing with android tts.

      The point is to figure a FOSS self hostable way for on the fly building of audio books from epubs and articles. And then, depending on online non-libre tooling feels a bit fragile.

      That one is Google’s with UK accent, and it’s far from perfect. I want to play with https://github.com/mozilla/TTS still, and see what comes out of it.

      In the meantime, I’ll also try to run balabolka on Linux and see if I can take something away from it. Thanks!