Record Your Screen With System and Mic Audio
You hit record, walk through the bug, narrate the fix — and then you watch it back and the video is silent where the app should be talking. The button click made a sound. The video you were reacting to had music. None of it is in the file. Just your voice, floating over a muted screen.
This problem has a specific cause: the recording captured your microphone but not the sound your computer was playing. This guide explains the difference between system audio and microphone audio, how loopback capture works on Windows, and how to record both at once — cleanly, in one file, without a single extra driver.
System audio vs. microphone: two completely different sources
When people say "record screen with audio," they usually mean two different things at once, and a recorder has to handle each separately:
- Microphone (input): your voice. This is the easy one. Every recorder can open a mic input. It's the same stream a voice call uses.
- System / desktop audio (output): everything the computer plays out — the app you're demoing, the YouTube video you're reacting to, a Zoom call, a game, a notification chime. This is the stream going to your speakers or headphones.
A microphone is an input device, and recorders are built to read inputs. System audio is an output — sound leaving the computer — and reading an output stream is a fundamentally different operation.
How capturing internal audio works on Windows
To record what your computer is playing, software has to do something called loopback: instead of listening to a microphone, it taps the audio device's render stream — the mixed-down signal on its way to the speakers — and copies it.
One route is a hardware loopback device. Microsoft's loopback recording documentation, read in October 2026, says some audio hardware vendors include one, under names such as Stereo Mix, and that the user must identify and enable it before an application can record from it. Another route is a virtual audio cable: software that adds a virtual audio device, so you route your system sound into it and then record that device.
The modern, correct answer is WASAPI loopback. WASAPI (Windows Audio Session API) lets an application open the existing output device in loopback mode and read the exact audio already being played — no virtual device, no rerouting, nothing for the user to configure. Microsoft's documentation says WASAPI supports loopback recording regardless of whether the audio hardware contains a loopback device.
Read Microsoft's loopback recording documentation.
How Maxisnap records both, driver-free
Maxisnap captures system audio through WASAPI loopback and your microphone at the same time, and by default mixes them into a single track so you hear everything — the app, the video, and your narration — together, the way a viewer expects. There's no Stereo Mix to enable, no virtual cable to install, no driver to chase down. You start recording and the sound is just there.
The recording itself is straightforward. Two hotkeys, both remappable:
Ctrl+Alt+4— record the full screen (the monitor under your cursor).Ctrl+Alt+3— record a selected region when you only want part of the screen.
A small recording badge appears so you always know it's rolling — and it's deliberately excluded from its own capture, so it never shows up in the video. To stop, press the hotkey again, click Stop on the badge, or use the system tray.
When you stop, a result window opens with an inline player, plus Save, Copy, Upload, and Drag-out buttons. You can trim the start and end, and export to MP4, AI-optimized MP4, GIF, or WebM with quality presets. Encoding can use hardware acceleration (NVENC, Intel QuickSync, or AMD AMF) when your machine supports it, with a software fallback otherwise — and ffmpeg ships inside the installer, so there's nothing else to download.
If you'd rather edit the voice and the system sound apart, you can keep them on separate tracks instead of mixing. That's the cleanest fix for echo and for cases where you want to duck the app audio under your narration later.
One track or two: how to choose
The default — system audio and microphone mixed into one track — is right for most people. It plays correctly everywhere with zero editing, and it's what you want for a quick demo, a tutorial, or a bug report.
Choose separate tracks when you plan to take the recording into an editor and want independent control: lowering the app's volume while you talk, removing a notification sound, or fixing a sync issue. Separate tracks cost you one extra step (you mix them down later) in exchange for full control.
Common failure modes — and how to fix them
| Symptom | Likely cause | Fix |
|---|---|---|
| Only my voice was recorded | Recorder captured the mic but not the system loopback | Use a recorder with WASAPI loopback (Maxisnap captures it by default — no setup) |
| No sound at all | Mic muted/wrong device, or output device changed mid-recording | Confirm the correct mic, and record the live output device directly via loopback rather than a virtual cable |
| System sound but no voice | Wrong microphone selected, or mic permission denied | Pick the right input device and grant microphone access in Windows privacy settings |
| Echo / doubled audio | Speakers play sound and the mic re-records it | Wear headphones, or record system audio and mic on separate tracks |
| Audio drifts out of sync | Frame rate or A/V timing mismatch during capture | Capture at a steady frame rate; trim and re-export from the result window if needed |
| Robotic / crackling sound | Sample-rate mismatch or an overloaded virtual-cable chain | Drop the virtual cable; loopback capture reads the device's native stream cleanly |
For Maxisnap, this table describes the Windows app. A Mac version is also available, for macOS 13 or later.
The thread running through almost all of these: every extra device in the audio path is one more thing to keep configured. Reading the output device directly through loopback removes most of the ways "record screen with internal audio and mic" can go wrong.
A quick note on Mac
On a Mac you can record the screen with Apple's Screenshot app or through QuickTime Player.
On earlier versions, app sound can be routed in through a virtual audio device such as BlackHole.
If you're on a Mac, the free Mac screen recording guide walks through exactly how to set that up. Maxisnap for Mac records system audio, on macOS 13 or later.
The fastest path to a recording that actually has sound
If you just want a clip with both the app audio and your voice in it, here's the whole flow on Windows:
- Press
Ctrl+Alt+4(full screen) orCtrl+Alt+3(region). - Talk over what you're demoing — system sound and your mic are both captured, mixed by default.
- Press the hotkey again to stop. Trim the dead air off the start and end.
- Export to MP4 — or pick AI-optimized MP4 (~1280px, 24fps, mono) to get a tiny file you can drop straight into ChatGPT, Claude, or Gemini.
No watermark, no time limit, and the recording is free. That's the difference between a silent video you have to re-shoot and a clip that says everything the first time.
Frequently asked questions
How do I record my screen with system audio and microphone at the same time?
Use a recorder that captures the system loopback as well as the mic. In Maxisnap, press Ctrl+Alt+4 (full screen) or Ctrl+Alt+3 (region); desktop audio via WASAPI loopback and your microphone are both captured and mixed into one track by default — no virtual cable or extra driver needed.
Why does my recording only capture the microphone and not the computer sound?
Because the recorder only opened the mic input. Capturing what the computer plays requires loopback recording of the audio device's output stream. Maxisnap does this through WASAPI, so desktop sound is captured driver-free alongside the mic.
Do I need Stereo Mix or a virtual audio cable on Windows?
No. Maxisnap reads the desktop loopback through WASAPI natively, so neither is required.
Why is there an echo in my recording?
The same sound is being captured twice — your speakers play it and your mic picks it up again. Wear headphones, or record system audio and the mic on separate tracks so you can mute or align one in editing.
Is recording with audio free, and is there a watermark or time limit?
Yes — recording with system audio and mic is free on Windows, with no watermark and no time limit. Local saving is free. Paid Cloud plans increase the hosted upload allowance; uploading from Maxisnap to your own server requires the separate Studio add-on.
Want the full walkthrough of recording without audio nuance? See how to record your screen on Windows for free, or learn how the combined screen recorder and screenshot tool fits into one lightweight app.
Maxisnap is not affiliated with Microsoft. This page describes Windows features as Microsoft's documentation presented them in October 2026; check Microsoft's documentation for current behavior.
Maxisnap is not affiliated with Apple. This page describes macOS features as Apple's documentation presented them in October 2026; check Apple's documentation for current behavior.
Maxisnap is not affiliated with the other products named on this page. They are described as their official sites presented them in October 2026; check each official site for current features, pricing and platforms.