A two-step chain built for voice recordings: reduce background noise using a sample you mark, then automatically bring the result up to a consistent, safe loudness.
Clean Voice Recording is the one-click chain for turning a rough voice recording — a phone memo, laptop-mic interview, lecture or voice-over — into something clean and consistent: rumble removed, background noise reduced, volume evened out and loudness set.
Each step is optional and adjustable, and a plain-language report tells you what was applied, in order.
Chain: 80 Hz high-pass (4th-order Butterworth) → spectral noise reduction (auto-learned profile, 0–100%) → optional soft gate (−48 dB threshold, −12 dB, 150 ms hold) → feed-forward compressor (−24 dB threshold, 1.5–3.5:1) → loudness normalisation to −16/−19/−14 LUFS with true-peak limiting. Noise reduction runs before compression so that the compressor does not raise the noise floor.
Make voice memos listenable.
Clean a classroom recording.
Fix remote interviews.
Prep voice-over.
| Chain | High-pass → noise reduction → gate → compression → loudness |
|---|---|
| Target loudness | −16, −19, −14 LUFS |
| Report | Every step listed |
Remove rumble, reduce noise, then compress — compressing first would amplify the noise.
No, classic DSP; it runs fully offline.
First it runs spectral-subtraction noise reduction using your marked noise sample (same engine as the standalone Audio Noise Reducer, at a medium strength), then it normalizes the result to a safe -1dB peak (same engine as the Audio Normalizer).
Not on this combined page — it uses sensible medium-strength defaults for both. For full manual control over each setting, use the Audio Noise Reducer and Audio Normalizer tools separately in sequence.
This tool runs entirely in your browser using the Web Audio API. Nothing is uploaded, stored or shared — there is no server to send your file to, so it works offline once the page has loaded.
Be the first to share your experience with this tool — your feedback helps others.