Audio Loudness Normalizer
Loudness normalisation aligns perceived level rather than peaks. For maximum volume, simple gain is the better tool.
Add files and press start. Everything runs locally in your browser; nothing is uploaded
How to Use
📖 Tool Introduction
Match the loudness of several audio files to one level, entirely in your browser. Recordings, downloaded stock and podcast episodes from different sources rarely arrive at the same level, and this tool makes them sit together.
This normalises loudness rather than peaks, which is not the same thing. A dynamic vocal track can be peaked to full scale and still sound quiet, while a compressed music track peaks lower and sounds louder. Loudness normalisation follows perception, so matching by ear is what actually comes out even.
Target loudness is the setting that matters. Minus 16 LUFS is the streaming platform norm and suits podcasts and spoken content; minus 14 is slightly louder and fits video soundtracks; minus 11 is louder still for noisy environments; minus 23 is the broadcast standard and keeps the most dynamics. The true peak ceiling normally stays at minus 1.5 dBTP so later transcoding cannot push it into clipping.
Normalisation aligns overall level and cannot rescue a bad recording. If the source has heavy noise or is already clipped, raising it amplifies those flaws along with the signal, and the fix belongs back at the original recording.
Processing uses a WebAssembly build of FFmpeg: about 24 MB downloads the first time and is cached after. Audio is never uploaded and stays on your device.
One more thing worth knowing before you start: normalisation changes level, not tone. A recording that sounds thin or boomy will still sound that way afterwards, only louder, and fixing the tone needs equalisation rather than this tool.
It also helps to know what the numbers mean in practice. LUFS describes average loudness as people hear it, so two files at the same LUFS sound equally loud even if one has far taller peaks, which is exactly the case with a compressed track next to a dynamic one.
✨ Key Features
- Loudness matching that follows perception rather than waveform peaks
- Four targets: -11, -14, -16 and -23 LUFS for different uses
- True peak ceiling from -1 to -3 dBTP, so later transcoding cannot clip
- Six output choices: keep source, MP3, WAV, FLAC, M4A, OGG
- Bitrate from 128 to 320 kbps, ignored for lossless formats
- Batch mode: add several files and each is normalised in turn
🚀 How to Use
- Drag audio into the drop area, or click to choose files
- Pick a target loudness; -16 LUFS suits podcasts and spoken content
- Leave the true peak ceiling at its default of -1.5 dBTP, and either keep the source format or pick another one
- Press start and watch the progress bar
- Press download when it finishes
⚠️ Things to Note
- Engine size: The first run downloads about 24 MB of WebAssembly engine, cached for later visits
- Loudness versus peaks: This normalises loudness, not peaks. A lower target keeps more dynamics and a very high one compresses them, which is why broadcast sits at -23 and streaming sits louder
- Cannot fix bad audio: Only overall level is adjusted. Noisy or already clipped source will have those flaws amplified, and belongs back at the original recording
- Lossless formats: Bitrate is ignored when WAV or FLAC is chosen, since the output is lossless
- Privacy: Audio is processed locally and never uploaded. Reloading the page clears it completely