By the ReScreen team · · 5 min read
The take was good. You explained the feature clearly, the demo behaved, and then, right on top of the one sentence that mattered, a notification chimed at full volume. Or the app's intro sound was twice as loud as your voice. Or you realise that the background music from a browser tab is under all twelve minutes. With a normal recorder the verdict is always the same: record it again. With separate audio tracks in your screen recording, it is a thirty-second fix.
What is wrong with one mixed audio track?
Most recorders that capture both your voice and the computer's sound combine them into a single track as they record. From that moment the two are one waveform.
That has consequences you only meet later:
- Turning your voice up turns everything up, including the sound effect that was already too loud.
- Noise reduction meant for your voice chews on the app's audio too, making music and interface sounds warble.
- A single bad moment on one source condemns both. The ping sits on top of your sentence, and no editor can lift it off.
- You cannot change your mind. Decide afterwards that the video works better with narration only, and there is no narration-only version to be had.
How ReScreen keeps audio apart
ReScreen records four independent tracks: screen, camera, microphone and system audio. Your voice and the Mac's own sound are captured separately, written separately and shown as separate Mic and System tracks in the editor. The details of how the app's sound is captured in the first place, with no loopback device, are in the guide to recording system audio on a Mac.
All incoming audio is resampled to 48 kHz, so a Bluetooth headset, a USB microphone and the system track line up without you thinking about sample rates. The microphone can be mono or stereo, with a live meter before you commit to a take.
In the editor, each track gets its own controls:
| Per-track control | Range |
|---|---|
| Mute | On / off |
| Volume | 0–200 % |
| Fade in | 0–5 s |
| Fade out | 0–5 s |
System audio starts at 60 %, so it sits under your narration without you touching anything. The whole mixing surface is described on the camera and audio page.
What can you fix afterwards with separate tracks?
A voice that is too quiet
Raise the microphone track on its own, up to 200 %, and leave the app's sound where it is. "Level voice volume" evens out the difference between the sentences you said leaning in and the ones you said leaning back.
App sound that fights your narration
Pull the System track down further, or mute it entirely. Your voice does not change. On a mixed track this repair is simply not available.
Room noise under your voice
"Reduce background noise" and "Voice enhancement" (Off, Light or Strong) apply a high-pass filter, a soft noise gate, levelling and compression to the microphone only. The product's audio passes through untouched, so music and interface sounds stay clean. This is conventional processing, not AI voice isolation, and it is honest about that; the step-by-step version is in how to fix bad audio in screen recordings.
A rough start or an abrupt end
Give either track a fade of up to five seconds. Fading the system audio out under your closing sentence sounds deliberate rather than chopped.
A stumble you want to cut
Audio repair and picture editing happen on the same timeline. "Remove Range" takes out the stumble across every track at once and closes the gap, so nothing slips out of sync. That side of the app is covered on the timeline editor page.
What separate tracks cannot fix
Separate capture is not magic, and it is worth knowing the limits before you rely on it.
If the app's sound plays through your speakers, your microphone hears it. It then lives on the mic track as well, and muting the System track will not remove it. ReScreen has no echo cancellation, so the answer is practical rather than clever: wear headphones when you record with system audio on.
Likewise, a dog barking in your room is on the mic track, under your words. The noise tools can reduce steady background sound; they will not surgically remove a bark from the middle of a sentence. Separate tracks protect you from the other source, not from your own room.
Who benefits most from separate mic and system audio?
- Course creators. A course is dozens of lessons recorded over weeks, in a room that does not sound the same twice. Being able to match voice levels across lessons without touching the software's audio is the difference between a consistent course and an uneven one. There is more on that workflow in screen recording for online courses.
- Product marketers and founders. Launch and demo videos often end up with music added elsewhere, or get repurposed as silent social clips. Muting one track and exporting again beats re-recording.
- Support agents. Record once with the product's sounds, export a narration-only version for a help article, and keep both from the same take.
- YouTube tutorial makers. A channel lives on consistent sound from video to video; the guide to building a YouTube channel for a software company explains why viewers forgive average picture long before they forgive bad audio.
The common thread: these are people whose recordings are long, unrepeatable or both. "Just record it again" is not a small request.
What file do you end up with?
Inside the project, the tracks stay separate for as long as the project exists, so you can come back next month and remix. Export renders your levels, mutes, fades and voice processing into a standard MP4 or MOV with ordinary audio that plays anywhere.
The ready-to-play MP4 written beside the project the moment you stop is the quick version: both sources mixed, no editing applied.
Next step: test it on a take you would normally bin
Record two minutes with your microphone and system audio both on, headphones in. Then, in the editor, mute each track in turn, drag the System track down, and switch on "Level voice volume".
ReScreen runs on macOS 14 Sonoma or later, on Apple Silicon and Intel, and every plan includes all four tracks and the full editor. Have a look at the pricing page, then download ReScreen for macOS and stop re-recording takes that only had one bad second in them.
Filed under