Find clipped, distorted samples in any audio file — with timestamps, severity and a clean/severe verdict.
AI-powered audio review & delivery for voice production teams
Drop your audio file here or click to browse
MP3, WAV, FLAC, AAC, M4A, OGG
This is a detector, not a repair tool.
Found clipping? Re-export at a lower level with the free Volume Normalizer. Want all six checks at once? Run the Voice Quality Analyzer, or check loudness with the LUFS Meter.
Upload audio to map every clipped passage with timestamps and severity
Audio clipping is what happens when a signal exceeds the maximum level a system can represent — 0 dBFS in digital audio. The peaks that should rise above the ceiling get sliced off flat, which adds harmonics that were never in the original sound. You hear it as harsh, crackly clipping sound on loud syllables, drum hits or laughter: the recording works fine at normal levels, then breaks up exactly where the waveform was flattened.
Clipping can enter your audio at several points: the microphone preamp (gain set too high), the interface's analog-to-digital converter, a plugin pushed past 0 dB inside your DAW, or a final export with too much makeup gain. Wherever it happens, the result in the file looks the same — runs of consecutive samples pinned at full scale, called flat tops.
The detector scans your file in 50-millisecond windows. A window is flagged when its peak reaches the digital ceiling (within 0.1 dB of full scale, or above it) and the samples show ceiling behavior — runs of identical at-peak samples, the signature of a flattened waveform. Adjacent flagged windows are merged into clipped segments with start and end timestamps.
Each segment is graded by clipped-sample density: MINOR for isolated single-sample peaks, MODERATE when at least 1% of the samples sit on the ceiling, and SEVERE for sustained flat-topping (10% or more). The whole file then gets one verdict — CLEAN, MINOR or SEVERE — so you can tell at a glance whether to ship it, fix it, or re-record it.
Prevention beats repair: once a waveform is flattened, the original peaks are gone. If the verdict is MINOR, re-exporting the project at a lower output level (or trimming the offending section) is usually enough. If it is SEVERE, the distortion is recorded into the take — lower your input gain so peaks land around −12 to −6 dBFS and record again. De-clipping plugins can soften the damage, but they reconstruct a guess, not the original audio, so for client deliverables a clean re-record is the safer call.
Clipping is distortion that occurs when a signal is louder than the maximum level the system can represent (0 dBFS in digital audio). The waveform's peaks are cut off flat, adding harsh harmonics you hear as crackle or breakup on loud passages. The detector finds those flattened peaks and timestamps every one.
Usually input gain set too high: a loud word, laugh or plosive overloads the mic preamp or the interface's converter and the top of the waveform is chopped off. Speaking closer to the mic than you sound-checked, or stacking gain across a chain (preamp + plugin + master), has the same effect. Aim for speech peaks around −12 to −6 dBFS while recording.
Only partially. The flattened peaks are information that was never recorded, so nothing can truly restore them. De-clipping tools interpolate a plausible peak and can make MINOR clipping less audible, but SEVERE, sustained flat-topping stays audibly damaged — re-recording at a lower input gain is the reliable fix.
Set input gain so your loudest delivery peaks around −12 to −6 dBFS, leave headroom in every plugin and bus, and put a true-peak limiter (ceiling −1 dBTP) at the end of the chain for safety. Then run your export through this detector before delivery — a CLEAN verdict means no sample ever hit the ceiling.
Distortion is any unwanted change to the waveform; clipping is one specific, common kind — the hard flattening of peaks at a level ceiling. Other distortion (a blown speaker, a bad cable, codec artifacts) can sound similar but shows no flat tops in the file. If this detector reports CLEAN yet you still hear grit, the problem is somewhere else in your chain.
VoiceDeck adds AI-powered audio & video review and delivery for your whole team — so every file ships in spec, automatically.
