Why turning it up can distort
Every digital audio file has a ceiling. Levels are measured in decibels relative to full scale, dBFS, and 0 dBFS is the loudest value a file can store. Turn a recording up and its loudest peaks reach that ceiling first. Anything that would go past it gets chopped flat, which is called clipping, and it sounds like crackle and harsh distortion on every loud word or drum hit.
So the trick to making audio louder without distortion is not simply more volume. It is managing the peaks, so the overall level can come up while the loudest moments stay under the ceiling.
Boost with a limiter
The quickest fix is a boost followed by a limiter. The boost raises everything by a fixed amount, and the limiter catches any peak that would cross the ceiling and turns just that moment down, fast enough that you do not hear it working. As a rule of thumb, a 6 dB boost doubles the signal level, and about 10 dB is what most people hear as twice as loud.
Limiting is not free. Each peak it catches loses a little punch, so a small boost is transparent while a huge boost on already loud audio squashes drums and makes voices sound flat and tiring. For a quiet voice memo or a lecture recorded from the back of the room, though, a solid boost is exactly the right tool.
Normalize to a loudness target
Normalizing is the smarter option when you care about consistency. Instead of adding a fixed amount, it measures how loud the audio actually sounds and adjusts it to hit a target. The measurement that matters is LUFS, loudness units relative to full scale. Unlike a peak meter, LUFS tracks loudness the way people hear it, averaged over the whole file.
Loudness normalization is the right choice for evening out podcast episodes, matching a set of voice notes, or checking how a song will sit on a streaming service. Done well, it applies one clean gain change to the whole file, so the natural rise and fall of the recording survives.
Do not confuse it with peak normalization, the older approach that raises a file until its single loudest peak touches the ceiling. One stray peak, like a cough or a snare hit, can stop the rest of the file from getting any louder, which is why two peak normalized files can still sound very different in volume.
The loudness streaming services play at
Most major platforms normalize playback so listeners do not have to keep reaching for the volume. Spotify plays music at -14 LUFS: it turns loud songs down to that level and turns quiet songs up, as far as their peaks allow. YouTube also targets -14 LUFS, but it only turns loud audio down, so a quiet upload simply stays quiet. Apple Music uses -16 LUFS for Sound Check, and Apple recommends -16 LUFS for podcasts.
Two lessons follow. First, making a song far louder than -14 LUFS gains you nothing on those platforms, because they turn it back down and you keep whatever damage the limiter did. Second, a quiet video or podcast really does need fixing before you upload it, because some platforms will not rescue it for you.
How to make audio louder for free
We built a free Volume Booster that does both jobs. It runs in the browser, nothing uploaded, we never see the file. Drop in an MP3, WAV, M4A, or even a video file, then pick a mode: normalize to -14 LUFS for music and streaming, normalize to -16 LUFS for podcasts and voice, or Boost +6 dB and Boost +10 dB for a quick lift with a limiter behind it.
Normalizing runs in two passes. The first measures your file, and the second applies the gain change while keeping true peaks at or below -1 dB. Listen to the preview, then download an MP3 or a WAV. If you dropped in a video, the download is its soundtrack, ready to put back under the picture in your video editor.
If the recording is noisy as well as quiet, clean it first with the Noise Reducer. Raising the volume raises the hiss too, so removing noise before you boost always gives the cleaner result.
When louder is not the answer
If a voice is buried under music rather than simply quiet, raising the whole file will not bring it forward. That is a balance problem, and it needs the voice and the music on separate tracks. And if a recording distorted while it was being made, because the input level was set too high, no tool can fully undo that clipping afterwards.
The lasting fix is at the source: record with peaks landing comfortably below the ceiling, somewhere around -12 to -6 dBFS, and make it loud afterwards. Quiet and clean is easy to fix. Loud and clipped is not.