RunLocalTools

Voice Recording Normalizer — Even Out Quiet and Loud Parts

Even out volume jumps in a voice recording with real compression, then export a leveled WAV or MP3. Free, no upload, no signup. Runs entirely in your browser.

What it does

Load a voice recording and this tool evens out the gap between the quiet and loud parts, so whispers don’t disappear and shouts don’t spike. You get a before-and-after waveform, can play the original and the processed version back to back, and then download the result as a WAV or MP3 file.

It suits podcasts, voice memos, interviews, lecture recordings, voiceovers and video narration.

How it works

Many “normalizers” just multiply the whole file by one number until the loudest peak hits a target. That makes a recording louder, but the quiet and loud parts stay the same distance apart, so a whisper is still a whisper.

This tool does two things in order:

  1. Dynamic range compression. Sounds above a threshold are turned down by a set ratio, and the result is brought back up with makeup gain. The loud parts come down toward the quiet parts, so the overall level is more even. Your browser’s built-in audio compressor does the processing, rendered offline so it’s faster than real time.
  2. Peak limiting. Afterwards the whole file is scaled so its loudest point sits at about −1 dB below full scale. Whatever makeup gain you choose, the export cannot clip.

Three presets cover most voices, and every setting stays editable:

  • Gentle: light levelling that keeps the natural feel.
  • Podcast (balanced): the default for talking-head recordings.
  • Aggressive: heavy levelling for very uneven or distant recordings.

An optional silence trim removes quiet stretches at the start and end. A frame only counts as silent when every channel is below your threshold, and a file that is entirely silent is never trimmed down to nothing.

How to use it

  1. Drop an audio file onto the tool, or click to choose one.
  2. Pick a preset. The threshold, ratio, attack, release, knee and makeup gain fill in so you can see exactly what it chose.
  3. Tweak any value if you like. The preset switches to “Custom” once you do.
  4. Tick Trim leading/trailing silence if you want dead air removed from each end.
  5. Choose WAV (lossless) or MP3 (smaller).
  6. Click Process, then compare with Play original and Play processed.
  7. Click Download result.

If your recording is in several pieces, join them with the Audio Merger first and level the combined file once, so the joins don’t change in volume.

Supported formats and limitations

  • Input: anything your browser can decode, which usually means MP3, WAV, M4A/AAC and OGG. If a format isn’t supported, you’ll see a clear message rather than a silent failure.
  • Output: 16-bit WAV, or MP3 at 160 kbps (the MP3 encoder downloads the first time you choose it).
  • This levels by compression and peak limiting. It does not measure loudness in LUFS, so it isn’t a substitute for broadcast loudness targets.
  • The whole file is decoded into memory, so a very long recording may be slow on a low-powered device.
  • Compression can raise background noise along with the quiet speech. If the noise floor is high, use a gentler preset.

Privacy

Your audio never leaves your device. It is decoded, processed and encoded in your browser. The file is kept in local browser storage only so it survives a refresh, and the Clear saved file button removes it.

FAQ

What’s the difference between normalizing and compressing? Normalizing scales the entire file by one amount so the peak hits a target. Compressing reduces the difference between loud and quiet parts. Levelling a voice recording needs compression; the final peak step then keeps the result from clipping.

Which preset should I use for a podcast? Start with Podcast (balanced). If it still sounds uneven, try Aggressive. If it sounds squashed or the background noise rises, go back to Gentle.

Why does my recording sound louder but still uneven with other tools? Those tools usually only raise the overall volume. Here the compressor narrows the gap between loud and quiet first, then the peak limiter sets the final level.

Will it clip or distort? No. After compression the file is scaled so its loudest point sits about 1 dB below full scale, regardless of your makeup gain setting.

Should I export WAV or MP3? Choose WAV for further editing, since it is lossless. Choose MP3 for sharing or uploading, since the file is much smaller.

Can it fix background noise? No. It evens out volume; it does not remove noise. Compression can make a constant hiss more noticeable, so clean noisy recordings first where you can.