Skip to main content
Edit Videos

Free · private · in your browser

Normalize video audio

Choose a video, compare its level, then create a balanced result.

How it works
  1. ImportChoose source
  2. EditSet the change
  3. PreviewReview result
  4. ExportCreate file
InputMP4, WebM, MOV and M4V videoOutputMP4 or WebM video

Normalising measures how loud a video actually is and adjusts it to a target, so viewers do not have to reach for the volume control between your clips and everyone else’s. This tool analyses the soundtrack, shows you where it currently sits, and produces a version at a consistent level — the standard final step before publishing anything.

Loudness is not peak level

The old approach — peak normalisation — finds the single loudest sample and scales everything so it just touches the maximum. It is a poor proxy for how loud something sounds, because one stray peak can hold back an otherwise quiet track. Loudness normalisation measures the perceived level across the whole programme instead, weighted roughly for how hearing works, and expressed in LUFS. Two clips normalised to the same LUFS value sound equally loud; two clips peak-normalised to the same value very often do not.

Targets that match where the video is going

Most platforms now normalise on their end, and knowing their target saves you from being adjusted twice. Around −14 LUFS is the common figure for online video, while broadcast standards sit closer to −23. Uploading something mastered far louder than the target does not make it louder to the viewer — the platform simply turns it down, and you are left with the dynamic range you crushed to get there. Normalising to the target yourself gives you the same final loudness with the dynamics intact.

Where normalising fits in the order of work

It belongs at the end. Noise reduction, equalisation and compression all change the level, so a track normalised first drifts off target as soon as you touch anything else. The sequence that works is: clean up the noise, shape the tone, control the dynamics, then normalise, and finally limit if you need protection against isolated peaks. Normalising a heavily clipped recording is the one case where it cannot help — raising the level of distorted audio just gives you louder distortion.

Questions, answered

What loudness should I target for online video?

Around −14 LUFS suits most online platforms, since it is roughly where they normalise anyway. Broadcast work targets nearer −23 LUFS.

What is the difference between normalising and just raising the volume?

Raising volume applies a fixed gain regardless of content. Normalising measures the perceived loudness first and applies exactly the gain needed to reach a target, so different clips end up matching.

Why does my video still sound quieter than others?

Usually because the others are more compressed. A track with a narrow dynamic range sits at a higher average loudness at the same peak level. Compression before normalising closes that gap.

Should I normalise before or after other audio edits?

After. Noise reduction, EQ and compression all change the level, so normalise last — followed by a limiter only if isolated peaks still need catching.

Can normalising fix distorted audio?

No. Clipping is baked into the recording, and increasing the level simply makes the distortion louder. Only re-recording fixes it.

Useful next steps

Continue with a tool that fits naturally after normalize video audio.

Next toolChange video volumePreview and automate volume changesNext toolVideo audio compressorPreview and balance quiet and loud momentsNext toolVideo audio limiterFind, preview and catch loud audio peaksNext toolDuck music under voiceLower background music during speech