Conversation

Notices

Adrian Cochrane (alcinnz@floss.social)'s status on Tuesday, 14-Feb-2023 05:35:38 JST Adrian Cochrane

There's numerous projects I am grateful for, but here's a few:
* FontConfig
* FriBidi
* Harfbuzz
* LibICU
* Haskell
* XML Conduit
* http-client-tls
* Haskell CSS Syntax
* Rasterific SVG
* Juicy Pixels
* Mesa3D
* Cryptonite
* eSpeak NG
* nginx
* Maddy
* Snikket/Prosody
* SourceHut
* Voice2JSON
Without these it'd be a lot harder to develop the #ArgonautStack !
#iLoveFreeSoftware

In conversation Tuesday, 14-Feb-2023 05:35:38 JST from floss.social permalink
- Adrian Cochrane (alcinnz@floss.social)'s status on Tuesday, 14-Feb-2023 06:13:30 JST Adrian Cochrane
  in reply to
  
  What are these libraries?
  FontConfig, Harfbuzz, & FreeType all handles fonts.
  FriBidi & LibICU handles unicode text. Together these all allow me to excel at internationalization!
  Haskell is a safe & elegant language I've come to love.
  XML Conduit is an XML parser, Haskell CSS Syntax is a CSS lexer. http-client-tls is self-explanatory, using Cryptonite for crypto.
  Rasterific SVG renders SVG, JuicyPixels other image formats.
  Mesa3D are linux's userspace GPU drivers, our OpenGL implementation.
  
  In conversation Tuesday, 14-Feb-2023 06:13:30 JST permalink
- Adrian Cochrane (alcinnz@floss.social)'s status on Tuesday, 14-Feb-2023 06:23:48 JST Adrian Cochrane
  in reply to
  
  eSpeak NG is a handwritten multilingual text-to-speech engine! Voice2JSON parses speech-to-text voice commands without involving the internet!
  nginx is the HTTP server I like using.
  Maddy & Snikket/Prosody are email & XMPP servers. Very easy to setup, & I like those communication protocols!
  SourceHut is a lightweight, flexible, & capable suite of software-project management services!
  3/3
  
  In conversation Tuesday, 14-Feb-2023 06:23:48 JST permalink
- Adrian Cochrane (alcinnz@floss.social)'s status on Tuesday, 14-Feb-2023 06:31:22 JST Adrian Cochrane
  in reply to
  - Haelwenn /элвэн/ :triskell:
  @lanodan Several!
  Mozilla DeepSpeech (is it still around?)
  PocketSphinx
  Kaldi
  Julius
  
  In conversation Tuesday, 14-Feb-2023 06:31:22 JST permalink
- Haelwenn /элвэн/ :triskell: (lanodan@queer.hacktivis.me)'s status on Tuesday, 14-Feb-2023 06:31:24 JST Haelwenn /элвэн/ :triskell:
  in reply to
  
  @alcinnz > speech-to-text voice commands
  
  Do you know of one doing generic speech-to-text btw? Use case is when subtitles are missing from a video.
  
  In conversation Tuesday, 14-Feb-2023 06:31:24 JST permalink
  
  Adrian Cochrane repeated this.
- MrMagne (mrmagne@framapiaf.org)'s status on Tuesday, 14-Feb-2023 15:20:44 JST MrMagne
  in reply to
  - Haelwenn /элвэн/ :triskell:
  @lanodan @alcinnz this post might be of some interest https://m.filmfreedom.net/notes/9b5xy5fjbc
  In conversation Tuesday, 14-Feb-2023 15:20:44 JST permalink
  Attachments
  1. Terry Hancock (Director/Producer) (@TerryHancock)
    
    I just found out about automatic captioning in Kdenlive last week, and this is really a nice feature! Was added in v21, using the Vosk speech recognition library and language models -- all open source and all local (self-contained on my workstation). Took about an hour to do a 1.8 hour video, even using the large language model, so that's basically realtime! Of course, it does make mistakes -- about as many as automatic captioning on YouTube, by my impression. This opens some doors for me. Supports a variety of languages, not just English (though I haven't tried the others). https://docs.kdenlive.org/en/effects_and_compositions/speech_to_text.html

Public

Conversation

Notices

Feeds