User Interface
Running voicepad (or uvx voicepad) opens the interactive terminal interface. It is built with Textual and runs entirely in your terminal.
Layout
The interface has a header bar at the top, three tabs in the middle, and a keyboard shortcut bar at the bottom.
Header Bar
The header shows the app name, version, current status, and the active Whisper model with its compute device (cuda or cpu).
| Status Colour | Meaning |
|---|---|
| Green | Ready to record |
| Red | Recording in progress |
| Yellow | Transcription running |
Keyboard Shortcuts
| Key | Action |
|---|---|
| Space | Start or stop recording |
| C | Copy the current transcription to clipboard |
| Q | Quit VoicePad |
| I | Open the info panel |
| Ctrl+P | Open the command palette |
| Tab | Switch between tabs |
| Esc | Close a modal or dismiss |
Tabs
Record
The main tab. Press Space to start recording. While recording, a live timer shows in the status bar. Text appears in real time via streaming transcription as you speak.
When you stop recording, the final transcription appears in the panel. Press C or use the copy button to copy the text to your clipboard.
History
Browse all past recordings. Select a recording from the left panel to view its full markdown transcription on the right.
The retranscribe button re-runs the Whisper model on the selected WAV file. This is useful when you switch to a more accurate model after recording.
Settings
Configure VoicePad without editing any files. Changes are saved to your global config file when you press Save.
You can change the following settings:
| Setting | Description |
|---|---|
| Recordings path | Where WAV files are saved |
| Markdown path | Where markdown transcriptions are saved |
| Transcription model | Whisper model to use |
| Input device | Microphone to record from |
See Configuration for more details on each setting.
Info Panel
Press I to open the info panel. It shows the version, privacy guarantees, and links to the GitHub repository and sponsor page.