record commands
Audio recording commands for capturing and transcribing voice input.
voicepad record start {#start}
Start recording audio from the configured input device.
Synopsis
voicepad record start [OPTIONS]
Description
Begins audio recording from your microphone. Press Ctrl+C to stop recording. By default, the recording is automatically transcribed after stopping unless --no-transcribe is used.
When VAD chunking is enabled (--vad), transcription happens in real-time as chunks are detected, and the markdown file updates live while recording continues.
Options
--prefix TEXT, -p TEXT
Custom filename prefix (overrides recording_prefix in config).
Default: Value from config file (typically recording)
Example:
voicepad record start --prefix team_meeting
# Creates: team_meeting_20260218_033000.wav
--duration FLOAT, -d FLOAT
Record for a fixed number of seconds, then stop automatically.
Default: None (record until Ctrl+C)
Example:
voicepad record start --duration 300
# Records for exactly 5 minutes
--transcribe / --no-transcribe
Enable or disable automatic transcription after recording.
Default: --transcribe (enabled)
Example:
# Record without transcribing
voicepad record start --no-transcribe
# Explicitly enable (redundant, but allowed)
voicepad record start --transcribe
VAD Mode
When --vad is enabled, transcription always happens during recording.
The --transcribe flag only affects post-recording transcription.
--vad / --no-vad
Enable or disable VAD-based chunking for smart audio splitting.
Default: --no-vad (disabled)
Example:
# Enable VAD chunking
voicepad record start --vad
# Explicitly disable (redundant)
voicepad record start --no-vad
See: VAD Chunking Guide
--min-chunk-duration FLOAT
Minimum duration (seconds) before allowing chunk splits (VAD mode only).
Default: 60.0 seconds
Range: 10.0 - 600.0 seconds
Example:
# Larger chunks for lectures
voicepad record start --vad --min-chunk-duration 120
# Smaller chunks for Q&A
voicepad record start --vad --min-chunk-duration 30
--vad-threshold FLOAT
Speech detection sensitivity threshold (VAD mode only).
Default: 0.5
Range: 0.0 - 1.0
Example:
# More sensitive (quiet environments)
voicepad record start --vad --vad-threshold 0.4
# Less sensitive (noisy environments)
voicepad record start --vad --vad-threshold 0.6
Examples
Basic Recording
voicepad record start
# Press Ctrl+C when done
Output:
data/recordings/recording_20260218_033000.wavdata/markdown/recording_20260218_033000.md
Meeting with VAD
voicepad record start \
--prefix standup_meeting \
--vad \
--min-chunk-duration 60
Output (live updates):
data/recordings/standup_meeting_20260218_093000.wavdata/markdown/standup_meeting_20260218_093000.md(updates in real-time)
Fixed Duration, Audio Only
voicepad record start \
--duration 180 \
--no-transcribe \
--prefix voice_memo
Output:
data/recordings/voice_memo_20260218_103000.wav(only)
Sensitive VAD for Quiet Room
voicepad record start \
--vad \
--vad-threshold 0.4 \
--min-chunk-duration 45
Output Files
Audio file:
- Format: WAV (16-bit PCM, 16kHz mono)
- Location:
{recordings_path}/{prefix}_{timestamp}.wav - Created: When recording stops (VAD mode) or immediately (non-VAD mode)
Transcription file:
- Format: Markdown with metadata
- Location:
{markdown_path}/{prefix}_{timestamp}.md - Created: After recording (non-VAD) or during recording (VAD mode)
Stopping Recording
Press Ctrl+C to stop recording:
^C
[INFO] Stopping recording...
[INFO] Waiting for chunk worker to complete...
[INFO] Saving 5 accumulated chunks...
[INFO] Recording saved to: data/recordings/recording_20260218_033000.wav
Exit Codes
0- Recording completed successfully1- Recording failed (device error, permission denied, etc.)130- Interrupted by user (Ctrl+C - this is normal)
voicepad record info {#info}
Display the current recording configuration.
Synopsis
voicepad record info
Description
Shows the active recording settings from your configuration file, including:
- Input device
- Output directories
- Filename prefix
- Number of existing recordings
Output Example
Current Recording Configuration
============================================================
Input device index: 2 (Microphone - High Definition Audio)
Recordings directory: data/recordings
Markdown directory: data/markdown
Filename prefix: recording
============================================================
[OK] Recordings directory exists
12 recording(s) found
Tip: Use 'voicepad config input' to view and configure audio devices
Exit Codes
0- Always succeeds
Common Workflows
Standard Meeting Recording
# 1. Check setup
voicepad record info
voicepad config system
# 2. Start recording with VAD
voicepad record start --prefix team_meeting --vad
# 3. Let it run, watch markdown file update
# 4. Press Ctrl+C when done
Quick Voice Memo
voicepad record start --prefix memo --duration 60 --no-transcribe
Optimize for Accuracy
# Use larger chunks, adjust sensitivity
voicepad record start \
--vad \
--min-chunk-duration 90 \
--vad-threshold 0.5 \
--prefix lecture
See Also
- VAD Chunking - Smart audio splitting
- Background Transcription - Real-time processing
- Configuration - Setting defaults in YAML
- VAD Settings Reference - Parameter details