Skip to content

Settings: Audio & Video

Three tabs: Audio, Video and Mixing. The page opens on Audio.

Note

The Mixing tab's settings do not turn up in the settings search. Open the tab directly to find them.

Audio

Output

Not shown on macOS.

Output device
Which speakers count as "system audio" when you record. The first entry hands the choice to your system's own default.
The system default by default.
On Linux, choosing a specific device needs pactl. Without it the row goes amber and you get the system default only.

Input

The rows here are listed in the order your voice actually passes through them, which is worth knowing if you are chasing a sound problem.

Input device
Which microphone to record. First entry follows your system default, which is the default here too.
Echo cancellation
Cancels speaker sound picked up by the mic.
On by default. This is what stops the thing you are recording being heard again through your own microphone.
Noise Suppression
Reduces background noise.
On by default. Fans, air conditioning, traffic.
Automatic Input Sensitivity
Controls how much sound your microphone records.
On by default. Decides for itself where the line between your voice and the room is.
Input Sensitivity Threshold
A live meter with a threshold you set by hand. Anything quieter than the line is treated as room noise.
Only shown when Automatic Input Sensitivity is off.
Automatic Gain Control
Lifts quiet speech into the ideal range and holds it there, without crossing into too-loud.
On by default. This is the one that means you do not have to lean into your microphone.
Advanced Voice Activity
Smarter detection of when you're speaking.
On by default.
Push to talk
Hold the mic button to talk instead of pressing to toggle.
Off by default. With it on, your microphone is live only while you hold the microphone key. This row is hidden on systems that cannot support it.
Microphone test
A Test Microphone button. It opens a window that says "Speak normally. All active input filters are applied for this session", and draws your voice as a live waveform. That last part matters: you are hearing the same processing a real recording gets, not a raw feed.

The waveform is banded so you can see where you land:

Band Meaning
Ideal Peaks Where you want your loudest moments to sit
Normal Comfortable speaking level
Too Loud Back off, or move away from the microphone
Filtered Out Quiet enough that the app is treating it as room noise

Video

Video

Frame rate
Frames per second, from 1 to 240. 30 by default.
Max bitrate
Roughly how much data per second the video is allowed, in Kbps. Higher looks better and makes bigger files. From 100 to 500000. 8000 by default.
Max resolution
A ceiling on the recorded size. Anything bigger is scaled down to fit; nothing is ever scaled up. The choices are Original, 360p, 480p, 720p, 1080p, 2K, 4K and Custom.
2K (2560x1440) by default.
Max width and Max height
Your own ceiling in pixels, from 2 to 8192. 1920 and 1080 by default. Only shown when Max resolution is set to Custom.
Encoder
Which piece of hardware turns your screen into a video file. The list is what your machine actually has, named after the chip, and the app picks the best one for you on the first launch. A note appears if you have no hardware encoder at all, in which case recording falls back to your processor and works fine, just with more effort.
Encoder quality preset
How hard the encoder works per frame. The choices depend on which encoder is active, and every list runs from fastest to best quality with a sensible middle marked as the default. If your encoder does not offer a choice, the row shows "Driver default" and does nothing.
Video codec
Auto (by resolution), H.264 (max compatibility) or HEVC (smaller files). Auto by default, which is the right answer unless you have a reason.
An amber note appears if your codec and resolution choices would force a downscale, so it never happens behind your back.

Experimental

GPU zero-copy capture
Performance setting to preprocess frames on the GPU instead of the CPU when available.
Off by default. Only offered when your setup can actually do it.
Benchmark monitor and Benchmark encoders
Encoders that appear in green can sustain your currently configured frame rate. Encoders that use fewer cores will leave more processing for other programs.
Pick a monitor, press Run benchmark, and the app tests each encoder on that screen's real resolution. Results list the frame rate each managed and roughly how many processor cores it used. Green means it kept up with the frame rate you have configured. Neither of these is a saved setting.

Mixing

Pause other media during preview editor
When you capture content that contains audio, this will attempt to pause other audio while editing. Paused audio will resume after editing is completed.
On by default.
Automatically duck system audio
Automatically reduces recorded system volume when speaking.
On by default. Keeps your narration on top of whatever the screen is playing.
Automatically sync audio with video
On by default. Measures and corrects the delay between what you hear and what you see, per recording. Leave this on.
Audio sync offset
+ms delays audio (if sound is ahead of video), -ms advances it.
A manual correction from -1000 to 1000 milliseconds. 0 by default, and only shown when automatic sync is off.