Knip automatisch de stiltes uit je audio en video.
Upload je MP4 of WAV, wij detecteren de lange pauzes en geven je in seconden een bestand terug dat 30 % korter is. Zonder software, zonder registratie voor je eerste bestand.
- Free analysis and preview
- Nothing to install, all in the browser
- Files deleted automatically
- Premiere, Resolve, Final Cut timelines
Video demo
Two pauses removed from a real video
The same footage and voice before and after. The AutoCut engine removed 3.3 seconds from a 14-second video.
Before · 14 s
After AutoCut · 10.8 s
Footage: cottonbro studio / Pexels. Synthetic demo voice. The person shown is not affiliated with AutoCut.
Three moves, not one more.
- 1
Drop
One or several audio or video files. Loudness is measured every 20 ms; a threshold is proposed from your noise floor.
- 2
Listen and adjust
Cuts appear on the waveform. Before / after playback, presets, instant sliders, click to restore a passage, drag to cut one.
- 3
Export
MP3, WAV, M4A or MP4 with fades at every cut and normalized loudness, or the cut list for Premiere, Resolve and Final Cut.
Measured, not promised.
One measurement per frame
RMS level is computed every 20 milliseconds over the whole file. Every setting recomputes the cuts in milliseconds, on your machine, with no server re-analysis.
A fade at every cut
Every kept segment starts and ends with a micro-fade. No click, no pop: the cut is visible in the edit, not audible.
Two-pass loudness normalization
EBU R128 measurement on the already-cut audio, then a linear correction: podcast (−16) or YouTube (−14) level, without pumping.
The engine is the same in the demo above, in the tool and on the server at render time; its rules are covered by automated tests.
A preset for every use
Onafhankelijke podcasters
Neem je aflevering in één take op. Uploaden, klikken, publiceren. 20 minuten montage bespaard per aflevering.
YouTube- & TikTok-makers
Voor vlogs en talking heads: een schone MP4 zonder ongemakkelijke pauzes, klaar voor Premiere of DaVinci.
Online trainers
Je opgenomen cursussen verdienen een pro-look. Knip de aarzelingen en pauzes weg zonder het geluid aan te raken.
Journalisten
Interviews zijn zelden perfect. Maak ze in 30 seconden schoon voordat je publiceert of transcribeert.
Features for your edit
Adjustable settings, before/after listening and exports available in the studio.
| Make Autocut | |
|---|---|
| Price | €9.90/mo · €59/yr · €14.90 once |
| No software to install | ● |
| Silences shortened (not just removed) | ● |
| Before / after listening, one-click fixes | ● |
| “Um / uh” fillers | ✓ (beta, validated one by one) |
| LUFS loudness normalization | ● |
| Premiere / Resolve / Final Cut timelines | ● |
| API | ● |
| Text-based editing, captions | — |
| French interface | ● |
The price of an afternoon of editing.
Free to see and hear what it does. Paid to get the whole file and the Pro features.
- 3 analyses a day, files up to 10 min
- Waveform, instant settings, before / after listening
- 60-second extract of the result, MP3 or MP4
- Resume by e-mail link
- Everything Pro for 30 days
- No subscription, nothing to cancel
- Files up to 3 h and 2 GB
- 7-day history
- Everything Pro, unlimited files
- No commitment
- Cancel in one click
- New features included
- Everything Pro, all year
- Full files, MP3 320, WAV, MP4 HQ
- Fillers, normalization, noise
- Batches + ZIP, NLE timelines, API
- Priority queue
Start free, then choose a pass for one project or a subscription for regular use.
If the service does not work on your file: fix or refund within 14 days · Cancel online from your workspace · Prices include VAT
Frequently asked questions
Are my files confidential?
Yes. They are processed on our own server (Paris), never sent to a third-party service, and deleted automatically: 2 h on free, 24 h for the original and 7 days for the render on Pro. You can delete them earlier with one click.
How are silences found?
Loudness is measured every 20 ms. A threshold is proposed from the file's noise floor and voice; anything under it for at least 0.7 s is a silence. Each silence is shortened to 300 ms (adjustable), not removed: speech keeps its rhythm.
Does it cut in the middle of a word?
Not at the proposed settings: a gap must last 0.7 s to count, and 150 ms are kept on each side. If you push the threshold up, listen to the preview: you can restore any cut with a click on the waveform.
What about “um”, “uh”, “like”?
Pro transcribes the file word by word on our server (whisper model, nothing leaves) and proposes every filler for cutting. It is in beta: you validate each proposal before export.
Is audio quality degraded?
The file is re-encoded once (MP3 192 kb/s on free; 320 kb/s, lossless WAV or M4A on Pro). Every cut carries an 8 ms fade to avoid clicks. For a master, take the WAV.
What do I get for free?
The full analysis, the waveform with cuts, before / after listening, the statistics and a 60-second extract of the result, cuts included. The whole file, fillers, normalization, batches, timelines and the API are on Pro.
Pass or subscription?
The 30-day pass (€14.90) is bought once, no subscription, and gives everything Pro for a month. The subscription (€9.90/month or €59/year) renews and cancels in one click from your workspace.
Can I get a timeline for my editor?
Yes (Pro): FCPXML for Final Cut and DaVinci Resolve, XML for Premiere Pro, EDL, OpenTimelineIO and CSV. Your rushes are not re-encoded; you relink the media on import.
Is there an API?
Yes, included in Pro: drop a file or a URL, pick a preset, get the rendered file and the timelines back. Documentation is on the API page.
The longest edit is the one you never started.
One minute to drop, listen and get a clean file back.
Bestand uploaden