Updated October 2026 · Live streaming audio
Echo cancellation removes sound from your speakers that returns into your microphone, while noise suppression reduces steady or background sound around your voice. They solve different problems and can interact. Enable echo cancellation when speakers are in the room, and use noise suppression lightly to avoid damaging the voice.
Calls and streams often offer two settings that sound similar and do different jobs: echo cancellation and noise suppression. Many people switch them both on and assume the result will be better. Sometimes it is, but the two can interact, and the wrong combination can make the voice sound distant, watery or choppy. Understanding what each one does makes the settings easier to choose.
This post explains the difference, the problems each solves, the way they can interact and how to test them before a call or stream.
Echo cancellation solves a feedback problem. Sound from your speakers, such as the voice of the person you are talking to, can travel back into your microphone and be sent to them again. They hear their own voice a moment late, which is confusing and tiring. Echo cancellation compares the sound going to the speakers with the sound arriving at the microphone, and it removes the part that is an echo.
The effect is strongest when the sound from the speakers is a clean signal that the system can recognise. It is less effective in a very reverberant room, or when the speakers are loud, and it can fail if the delay between the speakers and the microphone changes.
Noise suppression reduces sound that is not speech, such as a fan, typing, traffic or a steady hum. It estimates which parts of the signal are speech and which are background, and it reduces the background. Modern suppressors can be quite effective on steady noise, but they can also remove quiet consonants or make the voice sound thin when set too strongly.
Noise suppression works on the sound around the microphone. Echo cancellation works on the sound coming from the speakers. They are different problems, and they need different solutions.
Using both can help, but the order and the strength of each matter. An aggressive noise suppressor can make the echo cancellation work harder, and the two can produce artefacts when they fight over the same sound. Some systems apply echo cancellation first and noise suppression second, while others differ. Check how your software orders them, and test the combination.
A practical rule is to use echo cancellation whenever the speakers are in the room, and to add noise suppression only if the background is a real problem. Starting with the lightest settings that work is safer than the strongest settings available.
Too much noise suppression makes the voice watery, thin or robotic, and it can cut the ends of words. Too little echo cancellation leaves the other person hearing an echo of themselves. Echo cancellation that is working too hard can sound like the voice is breaking up or dropping in and out. Listen for these signs, and reduce the processing when you hear them.
Record a short test and play it back. Check the voice for artefacts and the room for background sound. If possible, make a test call with a friend and ask them to tell you what they hear. Test with the speakers at the volume you normally use, since echo changes with the level.
Write down the settings that work, with the app, the version, the device and the date. A log turns a lucky setup into a repeatable one, and it helps you recover quickly after an update changes the menus or the defaults. Include the microphone, the distance, the echo setting, the noise setting and any notes about what you heard during the test.
Update the log when you change anything, even a small thing. A single line in a notebook is enough, and it is far more useful than trying to remember what you did a month ago.
Processing behaviour varies by app, device and version, and menu names change. The principles are stable, but the exact settings will differ. A setting that works in one app may behave differently in another, so test in the app you use.
Processing works best when the signal is already clean. The Clean Streaming Mic describes live noise cancellation that feeds OBS, Discord and Zoom, and the Tunedzy Go page covers cleaning recordings. For the wider noise options in apps, read the app noise settings post in this series.
Learn your room noise, remove it live and feed OBS, Discord or Zoom. Runs in your browser, nothing is uploaded.
Usually not, since headphones keep the speaker sound out of the microphone.
Noise suppression may be set too strongly. Reduce it and test.
Yes, but start light and test. The two can interact and cause artefacts.
A good microphone position and headphones, then light processing only where you hear a problem.
