Audio Settings
Audio sources
Voice Recorder Forge can record from three sources:
| Source | Description | |--------|-------------| | Microphone | Your device's default input device only | | System Audio | Audio playing from other applications (calls, videos, music) | | Both | Microphone and system audio mixed together |
Meeting recordings default to Microphone only. To include system audio, ensure the required platform setup is complete (see below).
Platform-specific audio capture
Windows
System audio capture on Windows uses WASAPI (Windows Audio Session API), which is built into Windows 10 and 11. No additional software is required. The app can capture audio from any running application.
macOS
macOS does not expose system audio to third-party apps by default. To capture system audio you must install a virtual audio device:
- Install BlackHole (a free, open-source virtual audio driver).
- Create a Multi-Output Device in Audio MIDI Setup that routes your speakers and BlackHole together.
- Set BlackHole as the app's input source.
See macOS Setup for a step-by-step guide.
Sample rate and channels
The audio capture service accepts optional configuration for sample rate and channel count. These settings are passed through to the underlying audio capture pipeline:
| Parameter | Default | Notes | |-----------|---------|-------| | Sample rate | System default | Higher rates improve accuracy but increase file size | | Channels | System default | Mono is sufficient for speech; stereo for mixed audio |
These advanced parameters are not currently exposed in the Settings UI and require configuration at the developer level.
Microphone permissions
The app will request microphone access on first launch. If you denied the request, re-enable it in your OS settings:
- Windows: Settings → Privacy & Security → Microphone
- macOS: System Settings → Privacy & Security → Microphone
Audio file format
Recorded audio is saved as WebM. For meeting sessions, the file is stored at:
[UserData]/meetings/[meetingId]-[timestamp].webm
Dictation sessions use temporary files that are deleted after transcription completes.