Features
Voicepad provides powerful features for voice recording and transcription.
Core Features
Recording
High-quality audio capture from any input device with customizable settings.
- 16kHz mono WAV format
- Configurable input devices
- Fixed or unlimited duration
- Automatic file management
VAD Chunking
Smart audio splitting at natural speech boundaries using AI-powered voice detection.
- Real-time chunk detection
- Configurable minimum duration
- Adjustable sensitivity
- Natural pause detection
Background Transcription
Real-time transcription processing while recording continues.
- Live markdown updates
- Model caching for speed
- Thread-safe file writing
- Single merged output file
Transcription Models
Multiple Whisper model options for different accuracy/speed tradeoffs.
- Tiny to Large-v3 models
- English-only variants
- Distilled models
- GPU acceleration support
Feature Comparison
| Feature | Without VAD | With VAD |
|---|---|---|
| Transcription timing | After recording | During recording |
| Markdown updates | Once, at end | Live, per chunk |
| Wait time | Full transcription | ~0 seconds |
| Memory usage | Entire recording | Per-chunk buffer |
| Best for | Short recordings | Long recordings (10+ min) |
Quick Links
- Getting Started - First-time setup
- Configuration - Customize settings
- CLI Reference - Command usage