Viewers forgive shaky footage far more readily than bad audio. Hiss, a low-frequency hum, boomy room echo, or a voice that’s too quiet in one shot and too loud in the next — any of these makes otherwise good footage feel amateur, and it’s the fastest way to lose someone in the first three seconds. The good news is that most audio problems fall into a handful of categories, and fixing them is a matter of doing a few things in the right order.
Fix it in this order
Order matters, because each step changes what the next one is working with. Denoise before you set levels, and set levels before you add music.
1. Remove steady background noise
Constant sounds — air-conditioner hum, computer-fan whir, tape hiss, traffic drone — are the easiest to remove because they’re predictable. Modern noise reduction learns the profile of the steady noise and subtracts it while leaving speech intact. Do this first: it’s hard to judge loudness or add music over a hiss you haven’t dealt with yet. Go easy — over-aggressive denoising makes voices sound underwater and metallic. Reduce the noise until it’s no longer distracting, not until it’s clinically silent.
2. Tame room echo (reverb)
Echo is trickier than hiss because the reflections are the same sound as the voice, just delayed. You can reduce it, but you can’t fully remove it, so the real fix is at recording time: get the mic closer to the speaker, and add soft surfaces (curtains, a rug, even a jacket over a chair) to a bare room. If you’re stuck with an echoey take, gentle de-reverb plus a small volume reduction on the quietest tails helps more than cranking a single slider.
3. Level the loudness
This is the step most people skip and the one viewers notice most. If one clip was recorded close to the mic and another across the room, splicing them together forces the viewer to ride their volume knob. Loudness normalization brings everything to a consistent target so the whole video sits at one comfortable level — the back of the room is audible without the close-ups being harsh. Aim for consistency across clips first; absolute level second.
4. Add music — carefully
Music sets energy, but it competes with speech for the same space. Two rules keep it from burying your voice: keep the music noticeably quieter than the dialogue, and duck it (drop its volume a few decibels) whenever someone is talking. Instrumental tracks beat vocal tracks under a talking-head clip, because two voices at once is fatiguing. When you add music to a video, set the speech level first and bring the music up underneath it — never the other way around.
Record better so you fix less
Every minute spent on the recording saves ten in editing:
- Get the mic close. Distance is the single biggest cause of thin, echoey, noisy audio.
- Record in the quietest room you have, away from fans, fridges and windows.
- Capture a few seconds of “room tone” (silence) so noise reduction has a clean profile to learn from.
- Watch your input meter — aim for a healthy level that never clips into distortion, which can’t be undone.
Do it privately, in your browser
You don’t need a digital audio workstation or a subscription to do any of this. Plumely denoises and levels loudness on your own device — your footage is never uploaded — so you can clean up a clip and export it without it ever touching a server. Start with removing background noise, then level the loudness, then add music. In that order, most clips go from “recorded on a phone” to “sounds intentional” in a few minutes.