audio sent for transcription · word-level timing
frame-by-frame compare · shot by shot
Apple Vision — on this Mac, nothing uploaded
appears here
Podcast clip maker for Mac — long podcast in, vertical clips out.
Podcast Clip Maker reads the camera cuts your editor already made, frames every shot on the people in it, and reframes the whole thing to 9:16 — then hands you the keyframes, so the final cut is yours. No uploads. No cloud render queue.
Not out yet. The app is still being finished, so there is nothing to download here today. When it is ready, it will say so right here.
Automatic where it saves you hours.
Editable where it matters.
Clip tools promise magic and lock you out of the result. We make the first draft — cut-aware, caption-ready — and keep the framing open for you to change.
No footage upload
Your recording stays where it is. Reading the frames, the camera cuts, the faces, the reframe and the export all run on your Apple Silicon Mac — not a frame of video ever goes out. Two things do: the audio track, for transcription (to our servers in the EU, or straight to your own provider if you paste your own key — it is deleted as soon as the text comes back), and a short line of numbers about the picture, so the vertical layout can be built for your plan.
Follows the edit you already have
The director already cut to whoever was talking. The app reads those camera cuts and re-frames each shot on the spot — snapping at the cut, never sliding.
Every frame is editable
Auto-framing is a first draft, not a final answer. Each crop is a keyframe you can drag until it’s exactly right.
Made for two-person shows
When the wide shot holds two people, they arrive stacked — both hosts on screen, top and bottom. There is nothing to pick: the app reads the shot, so the layout follows the edit.
Three steps, one honest pipeline
Drop in your episode
The finished export — the same file you would put on YouTube, multicam already cut, one audio track. Nothing to prepare.
Pick your clips
The transcript is indexed once on your Mac, word by word — the captions come from it too. Read the episode as text instead of scrubbing it, and mark the ranges worth posting.
Nudge and export
Fix any framing with a keyframe drag, zoom in where you want it tighter, and render 9:16 with hardware encoding.
Your master files stay put
A multi-camera podcast is hundreds of gigabytes. Cloud clippers make you upload all of it and wait. Here the media work happens on your Mac: reading the frames, the cuts, the faces, the reframe and the export. Transcription is the one step that needs a server — the audio track goes up for it, and nothing else does. What else leaves is a table of numbers about the picture (where the camera cuts, where faces sit), which our server turns into your vertical layout: about 100 KB per hour, no image and no sound in it. Transcript text leaves only if you link an AI assistant yourself.
Built with EchoLab, a video-podcast studio in Brussels. Every feature is shaped and stress-tested on real client footage before it ships — not on demo clips.
Questions podcasters actually ask
What actually gets sent anywhere?
Two things. The audio track, for transcription: the app pulls a 16 kHz mono track out of your file and sends it to our servers in the EU, which pass it to our transcription provider and delete it as soon as the text comes back. Paste your own provider key instead and the audio goes straight there — the key never touches us. A table of numbers about the picture: the timecodes of the camera cuts and the rectangles where faces sit, about 100 KB per hour of video, no image and no sound in it, and the vertical layout comes back. The video itself is never sent. Transcript text goes to an AI assistant only if you link one yourself. Exporting is always local.
Do I have to upload my footage?
No — not a frame of video, ever. A 3-hour multi-camera shoot is hundreds of gigabytes, and none of it has to crawl up your connection before you can start cutting. What does travel is the audio track, and only for transcription: a 16 kHz mono track, roughly 115 MB per hour, deleted on our side as soon as the words come back.
Do I need an account and a connection?
Yes for both, at two moments: you sign in with your Google account, and the pass that builds a project's vertical layout makes a short round trip to our server. Importing, preview and export don't need the network. Transcription does — it runs on a server, not on your Mac.
What do I get for free?
The free plan opens the whole app — the editor, the keyframes, export without a watermark — on the first minutes of each episode. Pro lifts that to the full length. The exact number of free minutes is shown in the app.
Which Macs are supported?
Apple Silicon (M1 or newer) on macOS 14 or later. Intel Macs and Windows may follow later.
Does it handle two-person podcasts?
Yes — that is the case it is built for. When both hosts are on camera it stacks them one above the other in a single 9:16 clip, and it holds that layout for the whole shot instead of flipping between them.
Can I fix the framing if it gets it wrong?
Yes. The automatic pass is a first draft, never a final answer: the crop is exposed as keyframes you can nudge, and any piece can be moved or resized by hand in seconds.
What resolution does it export?
Vertical 9:16 at 1080×1920, encoded with Apple's hardware encoder.
How is this different from OpusClip and other cloud clippers?
Cloud tools make you upload the whole episode and meter you by minutes or credits. Here the heavy work runs on your Mac — no footage upload, no render queue, nothing to wait in line for — and the framing stays editable instead of being a black box. Only the layout arithmetic makes a short trip to our server, in numbers.
Is there a watermark?
No watermark on exports.