Here's how we scaled to 1M
#startup#growth
Paste a link. Get back a batch of vertical clips that are ready to post - framed on whoever is talking, captioned, described and tagged.
500 free credits every month · no card · or upload a file
You already made the thing. The part that stops most people is cutting it up - which is the part that runs here while you do something else.
Paste a YouTube, Twitch or Zoom link, or upload the file. Connect a channel and new episodes come in on their own.
Transcript for what was said, vision for what happened on screen. Moments get scored on both, so a laugh counts as much as a sentence.
Cropped to whoever is speaking, captioned word by word, titled, described and tagged. Post them from here or download the lot.
Working from a specific source? See turning a YouTube video into Shorts or clipping one into TikToks.
Most auto-clip tools score moments from text alone. SnapDub also watches and listens - facial expressions, emotion, on-screen action, and the sound itself - so it still finds the moment even where the transcript can't. Always running on the newest AI models available, not whatever shipped last year.
Real reactions - laughter, surprise, excitement - the beats that actually make someone stop scrolling.
Leaning in, pointing, pacing around - visual energy a transcript alone can't capture.
Even a wordless or low-speech stretch still scores correctly, instead of defaulting to flat or blank.
A laugh, a roar from the crowd, a crash - moments that land in the audio and never reach the transcript at all.
Most tools hand you one clip at a time: cut it, style the captions, place the logo, export - then do all of it again for the next one, and hope clip 14 still looks like clip 1.
Clip 30 matches clip 1 because they were rendered from the same template - not because you remembered what you did three hours ago.
Run three shows or three clients at a time, keep a template for each, and switching between them is a dropdown.
Drop finished clips into a folder and put the same look on all of them at once - useful when half your library was cut somewhere else. They cost no credits at all, because there is nothing to generate, and they join your posting schedule like anything else.
Any single clip can still be opened and changed - drag its in and out points, fix a word the transcript got wrong, swap the caption style, move an asset. The other twenty-nine stay as they were.
Automatic until you disagree with it. Open any clip and the decisions the machine made are all still editable - and changing one clip never means rebuilding the batch.
Every clip is rendered with spare footage on both sides, so dragging the in and out points finds real frames instead of running out of video.
Caption text is editable, which matters for names, products and anything a transcript reliably mangles. Switch the whole look while you are there.
Animated emoji, your own images and GIFs, or a video laid over the clip with its own volume - dropped where you want them, timed to the second, and remembered for next time.
Connect a channel, pick the times you want to post, and the loop closes: every new upload is cut to your template and the clips go out on their own. Most tools stop at producing the clips - this one posts them.
A connected YouTube or Twitch channel is checked for you. A new episode is clipped to your template without anybody pasting anything.
Set the times once. New clips fill your next free slots, mixing clips from different recordings so a week does not read as one video sliced up.
Posted from here, not downloaded and uploaded again six times. Connect the accounts once and the clip goes where you told it to.
More on the timing side: posting on a schedule.
A link, a file, or a channel that feeds itself. All three are first-class inputs, and the third one is the one people forget exists.
An export, a downloaded VOD, a raw recording straight off the camera. It goes to storage directly, never through our web server.
YouTube, Twitch or Zoom. We fetch it ourselves, so you are not downloading a file first just to upload it again.
Connect it once and every new upload is clipped to your template on its own, without you opening anything.
Write a sentence, get a vertical video back - then caption it, brand it and post it in the same place as everything else you make here.
An opener, a cutaway, a product beauty shot - the pieces that hold a clip together and that nobody has footage of lying around.
Not a separate tool with its own downloads folder. A generated video arrives as an ordinary clip: same editor, same captions, same posting, same schedule.
It comes out of the credits you already have. Eight seconds starts at 249 credits, and you see the price before you press the button.
Four, six or eight seconds, vertical, on any paid plan. Longer pieces are made by joining generated clips the same way you join your own.
The full short-form pipeline, powered by open models running on our own GPUs.
AI reads the whole video for real topics - not just pauses and punctuation - so every clip opens and closes on a complete idea, never mid-sentence.
Karaoke-style captions burned in, styled to be read on mute - 29 looks to pick from, including keyword highlighting that lifts the word that matters out of the line.
Optional AI dubbing in the speaker's own voice, with lip-sync to match the mouth to the new audio.
Type a topic and instantly find the clips about it - across everything the video actually says, not just the AI-written titles.
Every clip comes with an AI hook title, caption and hashtags - copy them straight into the post.
Every clip is reframed to vertical and exported ready to post - the frame follows whoever, or whatever, the shot is about.
The same second of the same recording, cut two ways. A fixed centre crop on the left, ours on the right - including the part most crops get wrong, when the shot changes and the subject is suddenly somebody else.
SnapDub reads what is actually said in a clip, finds stock footage that fits it, and shows you three. Nothing is added on its own - one wrong cutaway ruins a clip worse than none at all, so the looking is ours and the choosing stays yours.
Most auto-clip tools stop at "cut it shorter." Here's what's different.
| SnapDub | Typical auto-clip tools | |
|---|---|---|
| Moment detection | Reads the whole video for real topics, start to finish | Fixed-length windows, blind to what's actually said |
| Finding a clip later | Search by keyword across everything said | Scroll and guess from the title |
| Generating video | Prompt to vertical video, from the same credits | The bigger ones have it, priced separately |
| Dubbing | Clones the speaker's own voice, lip-synced Beta | Captions only, or a generic TTS voice |
| Team seats | Included on every plan, shared credit pool | Priced per extra seat |
| Pricing | Credits you spend how you want | Locked to a fixed minutes/month quota |
| Cost per hour of video | ≈€3.46/hr on Starter | Roughly double that on comparable plans |
| Publishing | Post straight to TikTok & YouTube, no re-upload | Download, then upload yourself |
Coming from somewhere else? OpusClip, Klap, Submagic and AutoClip compared on what they each actually charge for, or put your own month through the calculator.
500 free credits every month - no card required. Credits are spent on the work you run - more clips, dubbing and lip-sync cost more. How credits work →
A batch of vertical 9:16 clips, each one cut on a complete thought, cropped to whoever is speaking, with word-by-word captions burned in and a title, description and hashtags written for it. Post them from here or download the lot.
Long enough to make a coffee, not long enough to plan your evening around. You do not have to sit and watch it - the clips are waiting when you come back, and we email you when they are ready.
Yes, and this is the part most tools skip. Every clip is rendered with spare footage on both sides, so you can drag its in and out points, fix a word the transcript got wrong, switch between 29 caption styles, pick a look for the burned-in title, turn the clip's own audio down, or drop an image, a GIF or a whole video onto the frame - then rebuild just that clip, not the batch.
On the free tier, yes - a small mark in the corner. Every paid plan renders without it, and you can put your own logo there instead through the brand kit.
Open the clip in the editor and drag its in and out points - every clip is rendered with a buffer of extra footage on both sides for exactly that. You can also re-crop it, fix caption text, or swap the assets on frame, then rebuild just that one clip.
Yes, and it does not run out after one go: every account gets 500 credits, and they are topped back up to 500 every month - enough to cut and caption a video each time - with no card and no trial timer. Subscribe or top up when one a month stops being enough.
Credits are spent on the work you actually run. A job costs a base amount per source-minute plus a per-clip charge - captions are cheap, while optional translation, AI dubbing and lip-sync add more, because heavier work uses more GPU time. A typical 10-minute video cut into 6 captioned clips - with title, description & hashtags included free - is ~140 credits. Failed jobs are refunded, and you can top up anytime.
No. Upload once and SnapDub picks the best moments, captions and exports vertical clips - each with a ready-to-post title, description and hashtags. Optional translation, dubbing and lip-sync (beta) are there if you want them. You just review and post.
Dubbing is an optional beta feature. With voice-clone dubbing enabled, we match the speaker's voice in the target language - quality varies, so treat it as a bonus. It's off by default, and you can also use a clean AI voice.
Translation covers 16 major languages, with slang and tone preserved. Most of them also support optional AI dubbing (beta).
Your media is stored on encrypted object storage and processed only to produce your clips - never sold or used to train third-party models. See our Privacy Policy.
No. Campaigns pay, on their own terms and their own schedule, and they decide whether a submission counts. We are the tool you make the clips with, we do not run campaigns or find them for you, and we are not affiliated with any campaign marketplace. How clipping campaigns work.
Yes. Manage, switch or cancel your plan right from your billing page at any time; cancellation takes effect at the end of the current period.
Build the look once, point it at a source video, and get a batch back. Start free with 500 credits a month - no card required.
Start clipping free