Audio Issues

No speech detected

If Vorn Voice shows an empty transcription or says no speech was detected:

Check your microphone. Make sure the correct input device is selected in your system audio settings. Test it with another app to confirm it is working.
Speak louder or closer to the mic. Some models struggle with very quiet audio.
Turn off aggressive speech cleanup. If speech cleanup is set to Aggressive, it may be trimming quiet speech. Switch to Balanced or Off in audio settings.
Increase pre-roll. If you start speaking immediately when pressing the hotkey, the beginning may be missed. Increase pre-roll from the default 350ms to 500ms or higher.

Beginning cut off: Increase pre-roll in audio settings. This captures audio from before you press the hotkey.
End cut off: Increase post-roll in audio settings. This captures audio after you release the hotkey.

The defaults (350ms pre-roll, 80ms post-roll) work for most people, but you may need to increase them if you speak quickly.

Try a larger model. Vorn Standard is the default, but Vorn Pro or Vorn Max are more accurate for longer or more complex speech.
Try cloud transcription. Cloud models are generally more accurate than local models.
Reduce background noise. Close windows, move away from fans, or use a headset microphone.

If you experience a noticeable delay before recording starts:

Make sure Low-latency capture is enabled in audio settings.
If the issue persists, try restarting the app. The persistent audio stream may need to be re-initialized.

On Linux, speech cleanup uses SoX for audio processing. If SoX is not installed, speech cleanup modes other than Off will not work.

Install SoX:

# Ubuntu/Debian
sudo apt install sox

# Fedora
sudo dnf install sox

# Arch
sudo pacman -S sox