Whisper + Omarchy - hyprwhspr, fast & accurate native dictation

[goodroot]

Checkout hyprwhspr!

It provides native integration of Whisper into Omarchy.

hyprwhspr

It's a simple installer that should have everything good to go once you restart.

Happy to help get you running and fix any bugs you might find.

[hipsterusername]

Hell yes.

[shanekunz]

@goodroot hey, linux/arch noob here, I tried to install it with the .sh script the other day and it didn't work, I get the logo in the wayland bar, but it always says "no microphone available". Is there a need to try to "uninstall" that version? Should I try the yay -S hyprwhpsr command even though I already ran the .sh?

[goodroot]

@shanekunz Yo!

When you open audio settings, can you confirm that the mic is indeed set in input settings? In Omarchy, you can right click to activate.

---

Also these logs will help determine things are OK.

# Check service status
systemctl --user status hyprwhspr.service

# Check logs
journalctl --user -u hyprwhspr.service -f

---

Ensure you have logged out/in or restarted to get the permissions set!

---

If there is still no joy after this, please run the install script again and paste me the output. We can see if there's something obvious.

[shanekunz]

@goodroot

the top terminal is those log commands. The bottom one is after running yay -S hyprwhpsr. I also tried running ./scripts/omarchy.install.sh like I did the first time and it succeeded, but still says "no microphone available" after trying to start and stop the service. super-alt-d does nothing.

[goodroot]

@shanekunz Thanks for your patience. We'll get you going. 😃

From the logs, hyprwhspr is starting and then stopping.

First, let's stop the service:

systemctl --user stop hyprwhspr.service

Clean up & reinstall:

# Remove all traces of the manual install
sudo rm -rf /opt/hyprwhspr

Run the automated installer

./scripts/install-omarchy.sh

Then reload user units and enable + start:

systemctl --user daemon-reload
systemctl --user enable --now hyprwhspr.service

Now we will tail the logs:

journalctl --user-unit hyprwhspr.service -f

The output of the log command should show what we need to debug. We should see RDY in the way bar indicating that we're ready once this has been done. Of course, once we reinstall, we will need to log out and log in once so that the permissions take place. Even if the microphone was not detected, I would still expect the application to start and even play the sound indicating that the recording action has begun, and ceased.