Help & Support
How can we help you?
Most questions are already answered below. If yours isn’t, send us the details and we’ll take it from there.
Speed up your reply
The more of this the first message carries, the fewer replies it takes to get you sorted.
- ✓Update first. Most issues are already fixed in the latest build.
- ✓Tell us your setup. Which operating system you are on, and the app version.
- ✓Name the format. Which export you used, and the editor you are importing into.
- ✓Describe the problem. What you expected to happen, what happened instead, and the exact wording of any error.
- ✓Say when it happens. Every single time, or only with certain files, settings or projects.
- ✓Show it if you can. A short screen recording, uploaded to any cloud drive, lets us see what you are seeing — send the link alongside your description, not instead of it.
No need to wait for us
- →Lost your license key? Have it emailed back to you.
- →Need an invoice or receipt? Download it from your order.
- →Changing or cancelling a plan? Manage the subscription yourself.
Common Support Questions
A lot of issues are solved in seconds — check here before writing to us.
Then the file genuinely carries no audio stream — nearly always a screen recording captured without system audio or a microphone. Silence detection listens to sound, so there is nothing for it to measure. Record again with an audio input enabled. If the silent clip belongs inside a longer sequence, you can still add it alongside clips that do have sound: it keeps its full length on the timeline and simply contributes silence.
That is deliberate. Loudness is a property of the footage, so the threshold is calibrated per clip, and the slider always edits the clip the playhead is sitting in — a quiet phone recording next to a mastered download needs a different line drawn, and one shared threshold gets both wrong at once. Move the playhead into another clip and the slider follows it. Min Silence and Padding describe your editing taste rather than the footage, so those stay the same across the whole sequence.
A sequence of mixed clips has to land on one format, and the rules are fixed: the frame size comes from the first video clip, the frame rate is the highest standard rate among your clips, and the bit depth is the deepest of them. Clips that do not match are scaled and padded to fit, never cropped. So if a portrait clip is being letterboxed into a landscape frame, put the clip whose shape you want at the top — drag it there in the review list before importing, or in the Clips window (the ⧉ button in the transport bar).
That is the known limit of the EDL format on a project with more than one source file. Camera and download files rarely carry a timecode track, so every clip's source times start near 00:00:00:00 and overlap completely — a CMX3600 file then has nothing left to tell your sources apart except a reel name your editor may not even be reading. For multiple sources, export FCPXML or Premiere XML instead: both reference each media file by an explicit id, so the match is exact. With a single source file, EDL is perfectly safe.
They belong there. When your edit speeds up the silences, a time remap applied by Premiere itself would play the voice back at the wrong pitch — the XML format has no flag to prevent that — so the app renders the retimed audio itself and pins it to the picture. Keep those .wav copies in the same folder as the XML and leave their names alone. Delete them and the sped-up stretches come in silent.
Three things account for almost every failure. The email you enter must be the one you bought with — it is checked against the key itself, so any other address is refused. The key is case-sensitive, so paste it from the purchase email rather than retyping it. And a key covers 2 computers at once: if both slots are taken, open Account → License Details on the machine you no longer use and press "Deactivate this device", which frees the slot immediately. Activation also needs the internet once, so a VPN, a corporate proxy or a captive-portal Wi-Fi can block it. If none of that is it, write to us under "Licensing & Activation" with your order ID.
Three sliders decide this. Raise the Threshold if pauses are being missed, lower it if quiet syllables are being clipped. Raise Min Silence so short breaths between words are left alone. Raise Padding to keep more air on either side of every cut. One thing that looks wrong but is not: the threshold is measured on peak level rather than an average, so the auto-tuned value normally sits around −32 dB, not the −41 you may expect. The reset button restores the value auto-tuned for your file, not a factory default.
No. The track you choose in the timeline toolbar only decides what silence detection listens to. Every audio track in the file survives the export, each one cut at exactly the same points. So pick whichever track has the cleanest speech — usually the mic — and let the rest follow it. You can also tick several tracks at once; the picker applies your choice when you close it, so selecting three costs one re-analysis instead of three.
An EDL, FCPXML or XML file holds only your cuts — it points at the media where that media was when you analyzed it. Moving, renaming or copying the sources to another drive after exporting is what breaks the link, so keep them in place or relink inside your editor. If you used Convert & Export, keep the whole folder together: the converted media copies are written next to the timeline file on purpose, so the set travels as one.
Your source is variable frame rate — normal for phones, screen recorders and anything captured in a browser, where frames arrive when they arrive. A timeline document written against variable timing slides a little further out with every cut. When you export a timeline from such a file the app offers Convert & Export: it renders a steady-rate copy of the media and writes the timeline against that copy, which is what holds sync. A direct MP4 export conforms the rate for you already, and Tools → Frame Rate Converter fixes files up front.
Writing a finished MP4 is by far the heaviest thing the app does, because every kept frame is encoded again. If you are carrying on in an editor, export an EDL, FCPXML or XML instead — it is written almost instantly and your editor does the rest. 4K sources, HDR and variable-frame-rate conversion each add real time. Rendering to a local drive rather than a network share, and closing other heavy apps, usually gives the biggest gain. If a render fails, the app retries once on software encoding by itself; send us the message it shows if the second attempt fails too.
Check your system clock first. The dates inside a license are signed and cannot be edited, so the app also keeps a high-water mark of trusted time — and a computer whose clock has fallen behind that mark is treated as out of license until it can check in with our server. Correct the date, connect once, and it repairs itself. If you have just bought Pro instead: buying and activating are separate steps, so open Account → Activate License and paste the key from your confirmation email. And the free trial is one per computer — reinstalling or clearing app data re-asks the server, which hands back the same 14-day window.