Noise suppression filters steady background sound such as fans and hum out of your microphone signal, and echo cancellation removes sound from your speakers that the mic picks up again. In Audio Recorder – Voice Recorder both are switched on by default and both work on the microphone only.
Neither switch is a studio tool. They are the same real-time voice processing Chrome uses for calls, applied before the audio reaches the recorder. Knowing what each one does makes it easier to decide when to leave them alone and when to turn one off.
What noise suppression does
Noise suppression listens for sound that behaves like background noise: constant, broadband, and not shaped like speech. A laptop fan, an air conditioner, traffic hum, or the hiss of a cheap mic preamp all fit that pattern. The browser estimates that noise floor and lowers it while letting your voice through.
It handles steady noise well. It is much weaker on sudden or speech-like sounds: a door slam, a dog bark, keyboard clicks, or someone talking in the next room. Those often survive, sometimes with a slightly chopped edge.
The trade-off is tone. Aggressive filtering can make a voice sound thinner or a little "underwater", especially on quiet syllables and word endings. For spoken notes, lectures, and interviews that is rarely a problem. For singing or an instrument it usually is.
What echo cancellation does
Echo cancellation solves a narrower problem. If sound comes out of your speakers while the mic is open, the mic hears it again. The canceller compares what is being played with what the mic captures and subtracts the played part.
This matters most when you record with the Both source: your voice from the microphone plus the sound of a browser tab. The extension routes captured tab audio back to your speakers so you can still hear it. Without echo cancellation, a speaker setup would put a second, slightly delayed copy of the tab into your mic track. With headphones there is little to cancel, so the setting has less to do.
Where the switches are
- Click the extension icon to open the popup.
- Open Settings.
- Find Noise suppression and Echo cancellation. Both are on out of the box.
- Toggle either one, then press Record. The next take uses the new values.
Settings are locked while a recording is running, so you can't flip these in the middle of a take. Stop first, change the switch, then record again. That keeps one take consistent from start to finish.
They do not touch tab audio
Both options are microphone constraints. When the source is Tab audio, the extension records the tab's sound as Chrome delivers it, without voice filtering. That is the right behaviour: music, a lecture video, or a webinar should not be run through a speech filter. If you want the technical background, read how tab audio capture works.
When to turn them off
| Situation | Noise suppression | Echo cancellation |
|---|---|---|
| Voice memo or lecture notes in a normal room | On | On |
| Voice plus tab sound on speakers | On | On |
| Voice plus tab sound on headphones | On | Either |
| Singing, guitar, piano, any music | Off | Off, if you use headphones |
| Quiet room, good external mic | Try off | Off, if you use headphones |
| Capturing room ambience on purpose | Off | Off |
Music is the clearest case. Noise suppression treats sustained notes and soft decays as noise and trims them away, and a held chord can come out pumping. If you capture song ideas, turn it off and keep the room quiet instead. The guide for musicians covers that workflow.
My own habit with a decent USB mic in a quiet room: noise suppression off, echo cancellation on only when speakers are playing. The voice sounds more natural, and the occasional background sound is easier to deal with than a permanently filtered tone.
A quick A/B test
The fastest way to decide is to hear both versions of your own room.
- Choose Microphone as the source.
- Record 15 seconds of normal speech with noise suppression on.
- Stop, open Settings, turn noise suppression off.
- Record the same 15 seconds again.
- Play both takes from the recordings list with headphones.
Listen to the pauses between sentences, not just the words. If the "off" take has an audible hum in the gaps and the "on" take keeps your voice natural, leave it on. If the "on" take sounds watery and the room is already quiet, leave it off.
What these switches won't fix
Real-time filtering is a guardrail, not a repair shop. It won't remove reverb from a bare, echoey room, it won't level a voice that drifts close and far from the mic, and it won't undo clipping from a mic set too hot. The extension also has no noise-reduction editor or effects after the fact: the only edit is trimming.
So most of the gain comes from the source. Set a sensible microphone volume, sit closer to the mic, and pick a softer room. If a take still needs work later, export it as WAV and see how to clean up a recording in a desktop audio editor.