Like, I push a button or key combo and talk at my computer and it types what I talk.

If not, why?

I do not mind writing some code to get this working but my time is very limited right now.

all 5 comments

sorted by: hot top controversial new old
[–] 13 points 1 year ago* (last edited 1 year ago)

update:

https://github.com/mkiol/dsnote

https://flathub.org/apps/net.mkiol.SpeechNote

dsnote/speechnote (it's the same thing) seems to be the best gui option. Check out the doc https://github.com/mkiol/dsnote?tab=readme-ov-file#insert-into-active-window


old comment:

https://github.com/icholy/whisperd

my bad this uses the api, but I guess it can be easily modified to use a local whisper.cpp

some other links to check out:

https://github.com/fengwk/linux-stt-input

https://github.com/PhilippeRo/IBus-Speech-To-Text

https://github.com/Elleo/ibus-deepspeech

  • source
  • [–] 6 points 1 year ago*

    I've been following this project

    https://github.com/epicenter-so/epicenter/tree/main/apps/whispering

    They recently got a bunch of investment and hired some staff, and so they're developing fairly quickly. I saw an update about local Whisper support the other day

    Assuming it stays user friendly (currently open source - MIT), it might be a good option for you. Previously it was here:

    https://github.com/braden-w/whispering

  • source
  • I would like this tech as well

  • source
  • [–] 1 point 1 year ago

    https://github.com/cjpais/Handy

    I have this working, still need to make a pr for arch. btw

  • source
  • [+] -2 points 1 year ago