← Glossary

Glossary

System audio

The sound the Mac itself plays, from any app, captured as part of a screen recording. macOS records it only through the screen capture API, and not every recorder asks.

Audio and cameraRecording


System audio is whatever the Mac is playing: a video in a browser tab, a notification, the app being demonstrated making a sound, a call. It is distinct from the microphone, which hears the room, and a screen recording can carry either, both or neither.

Why it is awkward on a Mac

For most of macOS's history there was no way for an app to hear what the Mac was playing. QuickTime Player and the Cmd+Shift+5 toolbar still cannot: they record the screen and a microphone, and the sound from the speakers is not on offer. The workaround was a virtual audio device such as BlackHole, routed through Audio MIDI Setup so the recorder could listen to it as though it were a microphone. It works, and it is fiddly, and it is how most guides on the subject still read.

Since macOS 13 the screen capture API can hand an app the system's audio alongside the frames, and a recorder built on it can record the Mac's sound with nothing to install. Not every recorder asks for it. The post on recording a Mac with internal audio goes through which do.

Keep it on its own track

The useful version of system audio is a separate track from the microphone. Mixed together at record time, a notification chime over your sentence is in the recording for good. Kept apart, it can be turned down or dropped afterwards while your voice stays. Prequel records the two separately, and the Audio panel has an include switch and a level for each.

Related terms

Questions people ask

Does QuickTime record system audio?
No. QuickTime Player and the Cmd+Shift+5 toolbar record the screen and a microphone. Capturing the Mac's own sound with them needs a virtual audio device such as BlackHole routed through Audio MIDI Setup.
Can I record system audio and my voice as separate tracks?
With a recorder that keeps them apart, yes. Prequel records the microphone and system audio as two tracks, so each can be included, dropped or set to its own level in the editor afterwards.