Configuration Reference

Detailed documentation for all configuration settings.

Setting Categories

File Paths

Control where audio and transcription files are saved.

recordings_path - Audio output directory
markdown_path - Transcription output directory

Audio Settings

Configure microphone input and recording behavior.

input_device_index - Microphone selection
recording_prefix - Filename prefix

Transcription Settings

Choose model and processing device.

transcription_model - Whisper model selection
transcription_device - CPU/GPU selection
transcription_compute_type - Precision level

VAD Settings

Fine-tune voice activity detection and chunking.

vad_enabled - Enable/disable chunking
vad_min_chunk_duration - Minimum chunk size
vad_threshold - Speech detection sensitivity
vad_min_silence_duration_ms - Silence detection
vad_speech_pad_ms - Speech padding

Configuration File

All settings are defined in voicepad.yaml:

# Paths
recordings_path: data/recordings
markdown_path: data/markdown

# Audio
input_device_index: 2
recording_prefix: recording

# Transcription
transcription_model: medium
transcription_device: auto
transcription_compute_type: auto

# VAD
vad_enabled: true
vad_min_chunk_duration: 60.0
vad_threshold: 0.5
vad_min_silence_duration_ms: 1000
vad_speech_pad_ms: 400