How to boost audio volume online — free, no upload
- Step 1Open the Peak Volume Normalizer — Go to volume-normalizer. The FFmpeg WASM engine loads in the page — no install, no sign-in.
- Step 2Drop the file you want louder — Drag in a single MP3, WAV, FLAC, M4A, OGG, Opus, or video file. It loads straight into the tab; nothing uploads.
- Step 3Choose your ceiling — Set Target peak (dBFS). −1 (default) is the safe maximum; choose lower if the file feeds another process.
- Step 4Pick the output format — Keep the source format or choose MP3/WAV/FLAC/M4A. WAV/FLAC avoid an extra lossy generation.
- Step 5Run the boost — Click run. Pass one measures the peak; pass two applies the exact safe gain. Short clips finish in seconds.
- Step 6Download the louder file — Save the
-peaknormresult. If the boost surfaced noise, follow with ai-noise-reducer.
Safe boost vs blind boost
The same files under a measured (this tool) and a naive fixed-gain approach. The measured method never clips; the fixed method clips whenever peak + gain > 0.
| Measured peak | This tool (to −1) | Blind +6 dB | Verdict |
|---|---|---|---|
| −12 dBFS | +11.00 dB → −1 dBFS | −6 dBFS (under-boosted) | Measured wins on level. |
| −6 dBFS | +5.00 dB → −1 dBFS | 0 dBFS (at full scale) | Measured safer. |
| −3 dBFS | +2.00 dB → −1 dBFS | +3 dBFS → CLIPPED | Blind boost distorts. |
| −1 dBFS | 0.00 dB (already there) | +5 dBFS → CLIPPED | Blind boost destroys peaks. |
Picking the right ceiling
All values are selectable on the −12 to 0 dBFS slider. Lower = more headroom; higher = louder but riskier on playback.
| Ceiling | Use it when | Note |
|---|---|---|
| −1 dBFS (default) | You want it as loud as safely possible | Standard consumer ceiling; minimal clip risk. |
| −3 dBFS | The file feeds a mix or another encode | Leaves 3 dB headroom for downstream gain. |
| −6 dBFS | Boosting but keeping clear safety margin | Conservative; good before heavy processing. |
| 0 dBFS | Absolute maximum and you accept the risk | Inter-sample overs possible — use true-peak-limiter for safety. |
Cookbook
Real boosts. Each shows the measured peak and the exact gain — the headline is that the gain is always chosen so the result stops at your ceiling, never past it.
Quiet podcast clip boosted to −1
A clip peaking at −10 dBFS gets +9 dB — loud and clean, no clipping.
Pass 1 (volumedetect): max_volume: -10.0 dB Target peak: -1 dBFS Gain applied = -1 - (-10.0) = +9.00 dB Result: peak at -1.0 dBFS.
An already-loud file the tool refuses to over-boost
A file peaking at −0.5 dBFS asked for a louder result actually gets pulled down to −1. The tool will not manufacture distortion.
Pass 1: max_volume: -0.5 dB Target: -1 dBFS Gain applied = -1 - (-0.5) = -0.50 dB Result: peak at -1.0 dBFS (slightly quieter, not clipped).
Boost and keep it lossless
Choose WAV output so the louder file has no extra MP3 generation loss.
Input: clip.wav (max_volume: -8.0 dB) Target: -1 dBFS -> gain +7.00 dB Output: WAV (pcm_s16le) Output: clip-peaknorm.wav
Conservative boost to −3 before mixing
If the boosted file goes into a project, leave headroom by targeting −3 instead of −1.
Source peak: -13 dB Target peak: -3 dBFS Gain = -3 - (-13) = +10.00 dB Result: peak at -3.0 dBFS, 3 dB headroom for the bus.
Boost exposed hiss — chain the denoiser
A big boost on a noisy source raises the floor. Boost first, then clean.
Step 1 (this tool): -22 dB -> +21 dB -> -1 dBFS (noise floor up 21 dB) Step 2: /audio-tools/ai-noise-reducer Result: loud and clean.
Edge cases and what actually happens
File already at full scale
By designIf the peak is already at or above your target, the computed gain is zero or negative, so the file is left as-is or turned down. The tool will not boost a maxed file into clipping.
Asking for more level than the file can give
By designPeak boost is capped by the file's loudest sample — once that hits the ceiling, you cannot go louder without distortion. If it is still too quiet perceptually, that is a compression/loudness job, not a gain job.
Boost raises background noise
ExpectedGain lifts signal and noise equally. A large boost on a noisy file makes hiss audible. Follow with ai-noise-reducer.
Target 0 dBFS chosen
SupportedAllowed but not recommended — playback reconstruction can exceed 0 dBTP. Use −1, or run true-peak-limiter for a true-peak guarantee.
Source already clipped
PreservedBoosting cannot repair clipping baked into the recording. It re-scales level only; the flattened peaks remain. Re-record with more headroom.
Boosting a lossy file
Re-encodedKeeping an MP3/M4A output re-encodes once more. Export WAV/FLAC to avoid stacking lossy generations.
Want every file the same loudness, not just louder
By designPeak boost does not equalise perceived loudness across files. Use loudness-normalizer for LUFS matching.
File over Free limits
413 too largeFree caps at 50 MB / 30 min / 1 file. Pro: 200 MB / 120 min / 10 files. Pro-media and Developer: 100 GB / unlimited duration.
Several files to boost
Single file onlyThe tool boosts one file per run (acceptsMultiple is false). Process each separately.
Re-boosting the output
PreservedRe-running on an already-boosted file re-measures the at-target peak and applies ~0 dB. No compounding, no damage.
Frequently asked questions
How is this safer than a normal volume booster?
It measures the file's true peak first, then applies only the gain needed to reach your ceiling. A blind fixed-gain booster can push an already-loud file past 0 dBFS and clip; this cannot.
How much louder can it make my file?
Up to the point where the loudest sample reaches your target (default −1 dBFS). A file peaking at −10 dBFS can rise +9 dB. It cannot exceed the file's own loudest moment without clipping.
Will boosting clip my audio?
No, not at or below your chosen ceiling — the gain lands the existing peak exactly on target. Targeting below 0 (the default −1) keeps a safety margin.
Why didn't my already-loud file get any louder?
Because its peak was already at or above your target, so there was no safe gain left to apply — it may even have been turned down slightly to meet the ceiling.
It is still not loud enough — what now?
Peak boost is limited by your loudest sample. To raise perceived loudness further you need compression (speech-leveler) or loudness normalisation (loudness-normalizer).
Does the boost add distortion or pumping?
No. It is a pure linear volume gain — no maximiser, no compression. The only audible change is level.
Will the boost make background noise louder?
Yes — gain lifts everything equally. If hiss becomes audible, run ai-noise-reducer afterwards.
Is my audio uploaded?
No. The boost runs in your browser via FFmpeg WebAssembly; the file never leaves your device.
What ceiling should I pick?
−1 dBFS for the loudest safe result, −3 dBFS if the file feeds further processing. 0 dBFS is allowed but risks inter-sample overs on playback.
Which formats can I export?
MP3 (libmp3lame), WAV (pcm_s16le), FLAC (flac), or M4A/AAC. WAV/FLAC avoid an extra lossy re-encode.
Can it boost multiple files at once?
No — it is single-file. Boost each file in its own run.
What are the limits?
Free: 50 MB / 30 min / 1 file. Pro: 200 MB / 120 min / 10 files. Pro-media and Developer: 100 GB and unlimited duration.
Privacy first
Every JAD Audio tool runs entirely in your browser via FFmpeg (WebAssembly) and RNNoise. Your audio files never leave your device — verified by zero outbound network requests during processing.