A simple desktop app (PyQt6) for transcribing audio/video with Whisper
(via faster-whisper) - multi-file queue, GPU support, TXT/SRT output,
and a sleep-prevention guard while it works.
Two versions are available, one per operating system. Pick your system below.
Tested on Debian with KDE Plasma.
cd linux
bash install.shFull instructions: linux/README.md
cd windows
then double-click install.bat (or read windows/START_HERE.txt
for the short version).
Full instructions: windows/README.md
| Feature | Details |
|---|---|
| Multi-file queue | Add several files at once, processed in order |
| Interface language | English (default) / Polish, switchable anytime |
| Whisper model | tiny / base / small / medium / large-v3 |
| Device | Auto / CPU / GPU (CUDA) |
| Model download | Real progress: percent, size, speed, ETA |
| Transcription progress | Percent, ETA, speed, live log |
| Pause / resume | Available mid-transcription |
| Output | TXT and SRT, saved to Desktop by default |
| Sleep prevention | Blocks sleep/hibernation while running |
Current version: 0.3
Not specified yet.