Does it show the loudness before and after?

Yes. After normalizing, a line under the player shows the measured loudness before and after in LUFS and the gain applied in dB. The "after" figure is measured on the saved file, not guessed.

Is this the same as just making everything as loud as possible?

No — this targets a broadcast-standard loudness level (EBU R128, the same standard used by broadcasters and streaming platforms), which is different from simple peak normalization. It's built to make speech and music sound consistently loud without clipping or feeling overly compressed.

What is the difference between LUFS and peak normalization?

LUFS targets perceived loudness over the whole file, which is what streaming and broadcast standards use. Peak normalization only raises the loudest instant to a ceiling, so a file with one sharp spike can still sound quiet. The first four targets use LUFS; Peak to -1 dB is the peak method.

Which target should I choose for a podcast?

Podcast (-16 LUFS) uses a peak ceiling of -1.5 dB. Spotify normalises to -14 LUFS, so pick Spotify / streaming (-14 LUFS) for that; the YouTube option uses the same -14 LUFS as a common streaming reference (YouTube publishes no official target). Broadcast is -23 LUFS. Audiobook (ACX) aims at about -20 dB RMS (LUFS used as a proxy) with peaks held to -3 dB. Check the requirements of the place you are publishing.

Why did my file end up quieter than the target?

Reaching the target would have pushed loud peaks past the ceiling for that preset, so the tool stops the gain at the safe limit and says so in the summary. A single constant gain cannot fix this; reducing the peaks needs a limiter or compressor.

Does it change the sound quality or the format?

It changes only the volume, with no compression or equalizer. The file is re-encoded in the same family as the original, so a WAV stays lossless while an MP3 is encoded again once and may lose a trace of quality.

Can it normalize several files together so they match?

It handles one file at a time. Running each file through the same target gives them similar loudness, though they are measured and adjusted separately, so tracks with very different peaks can still end up a little apart.

Home/Audio Tools/Audio Volume Normalizer

Normalize Audio Loudness (LUFS) Online

Even out volume levels across an audio file

Files never storedRuns on your device

Your files are processed on your device. The first use downloads a processing engine, which your browser then keeps.

Drop an audio fileTap to choose an audio file

or click to browseor tap to choose

Even out volume levels

Large files (over 500 MB) may take longer or run out of browser memory

About Audio Volume Normalizer

Raises or lowers the whole file by one constant amount so its overall loudness reaches a target, using the same loudness standard (EBU R128) broadcasters and streaming platforms target. Useful for a recording where volume drifts up and down, or before uploading to a platform with its own loudness target. Runs entirely in your browser — the audio never leaves your device.

Watch: Audio Volume Normalizer in Action

Watch on YouTube · Subscribe for new tutorials

Is my audio uploaded?: No. Measuring and changing the volume both run in your browser, so the recording is not uploaded. The only download is the audio engine, about 25 MB, the first time an audio or video tool needs it.
Does it show the loudness before and after?: Yes. After normalizing, a line under the player shows the measured loudness before and after in LUFS and the gain applied in dB. The "after" figure is measured on the saved file, not guessed.
Is this the same as just making everything as loud as possible?: No — this targets a broadcast-standard loudness level (EBU R128, the same standard used by broadcasters and streaming platforms), which is different from simple peak normalization. It's built to make speech and music sound consistently loud without clipping or feeling overly compressed.
What is the difference between LUFS and peak normalization?: LUFS targets perceived loudness over the whole file, which is what streaming and broadcast standards use. Peak normalization only raises the loudest instant to a ceiling, so a file with one sharp spike can still sound quiet. The first four targets use LUFS; Peak to -1 dB is the peak method.
Which target should I choose for a podcast?: Podcast (-16 LUFS) uses a peak ceiling of -1.5 dB. Spotify normalises to -14 LUFS, so pick Spotify / streaming (-14 LUFS) for that; the YouTube option uses the same -14 LUFS as a common streaming reference (YouTube publishes no official target). Broadcast is -23 LUFS. Audiobook (ACX) aims at about -20 dB RMS (LUFS used as a proxy) with peaks held to -3 dB. Check the requirements of the place you are publishing.
Why did my file end up quieter than the target?: Reaching the target would have pushed loud peaks past the ceiling for that preset, so the tool stops the gain at the safe limit and says so in the summary. A single constant gain cannot fix this; reducing the peaks needs a limiter or compressor.
Does it change the sound quality or the format?: It changes only the volume, with no compression or equalizer. The file is re-encoded in the same family as the original, so a WAV stays lossless while an MP3 is encoded again once and may lose a trace of quality.
Can it normalize several files together so they match?: It handles one file at a time. Running each file through the same target gives them similar loudness, though they are measured and adjusted separately, so tracks with very different peaks can still end up a little apart.

How to use Audio Volume Normalizer

Drop one audio file, then pick a Target. The six buttons are Podcast (-16 LUFS), Spotify / streaming (-14 LUFS), YouTube (the same -14 LUFS as a common streaming reference, since YouTube publishes no official target), Broadcast (EBU R128) at -23 LUFS, Audiobook (ACX) at about -20 dB RMS (LUFS used as a proxy) with a peak limit of -3 dB, and Peak to -1 dB, which only makes the loudest point reach -1 dB.

Press Normalize Volume. The tool measures the file, applies one fixed gain, then measures the saved result again and adjusts once if it missed. A line below the player reports the loudness before and after and the gain applied, for example plus or minus a number of dB. Listen to the preview, then download the file, whose name starts with normalized_.

When to use it

Reach for it when a recording needs to meet a platform's loudness target before upload, when several voice tracks sit at clearly different levels and you want them to match, or when a quiet phone recording needs a measured boost instead of guesswork. The Audiobook target is there for submissions that specify an ACX-style level.

When not to use it

It applies one constant gain to the whole file, so it does not even out loud and quiet moments inside a recording; a compressor or a leveller in an audio editor does that. It will also lift background noise along with speech, so clean noisy takes with the Noise Remover first or expect hiss to rise.

Tips for best results

When a target cannot be reached, the summary tells you that loud peaks stopped the gain, and the result is deliberately left a little quieter rather than clipped. Pick Peak to -1 dB when you only want maximum headroom without a loudness goal. If the summary warns about a big boost, treat that as a hint to re-record or denoise instead.

Limitations

The gain is a single value for the whole file, so the difference between its loud and quiet parts is unchanged. When peaks limit the gain, the final loudness lands below the target. Output keeps the input's family: WAV stays WAV, FLAC stays FLAC, M4A, AAC and MP4 become M4A, OGG and Opus become OGG, and everything else becomes MP3, so lossy files are re-encoded once.

Privacy

Measuring and gain-changing both run in an engine inside your browser tab, so the recording is never transmitted. The only download is that engine, about 25 MB, pulled from a public CDN the first time any audio or video tool needs it. The adjusted copy is built in memory and handed to you as a download; the original on your disk is never modified.

Peak vs loudness normalization

Peak normalization raises the loudest instant to a ceiling, so one sharp spike decides how loud the file gets and the rest can still sound quiet. Loudness (LUFS) normalization measures how loud the whole file sounds and moves it to a target, which is what streaming and broadcast standards use. Five of the six targets here are loudness targets; Peak to -1 dB is the peak method.

Which LUFS target?

Pick the target that matches where the file will be published, and check that place's own requirements.

Normalization targets in this tool
Target	Loudness	Peak ceiling	Use it for
Podcast	-16 LUFS	-1.5 dB	Podcast episodes and voice recordings
Spotify / streaming	-14 LUFS	-1 dB	Music and anything for Spotify-style streaming
YouTube	-14 LUFS	-1 dB	The same level as streaming; YouTube publishes no official target
Broadcast (EBU R128)	-23 LUFS	-1 dB	Radio and TV delivery that asks for EBU R128
Audiobook (ACX)	About -20 dB RMS (LUFS used as a proxy)	-3 dB	ACX-style submissions; saved as a 192 kbps CBR MP3
Peak to -1 dB	No loudness goal	-1 dB	Maximum headroom when you only want it louder

How the loudness is measured

The tool measures your file, applies one fixed gain, then measures the saved file again and adjusts once if it landed more than 0.25 LU from the target. The result line shows the loudness before and after and the gain in dB. If loud peaks stop the gain, it says so, and the option to limit loud peaks turns the volume up as far as the target needs while catching the few peaks that would clip.

Normalize without uploading

Measuring and changing the volume both run in your browser tab, so the recording is not uploaded. The only download is the audio engine, about 25 MB, the first time an audio or video tool needs it.