Start here How it works

From a long video to a vertical clip

faster-whisper transcribes the video with word-level timestamps, PySceneDetect finds the scene boundaries, and Google Gemini 3.1 Flash-Lite scores the transcript to select the 3 to 15 strongest moments of 15 to 60 seconds each. Each moment is then cut with FFmpeg and reframed to 9:16 with MediaPipe face tracking.

Two stages decide whether a clip is usable: moment scoring and reframing. Everything else is mechanical.

Clipping AI Cloud runs the pipeline on a GPU from $12/month, with the first video free up to 60 minutes.

Give it a try Drop in a link, receive vertical clips

Your first video is free up to 60 minutes, then you get 20 free minutes a month, with no card requested.

Dashboard Your first video free up to 60 min · no card needed · no watermark from $12/month

What is Clipping AI?

Clipping AI is an AI clip maker that turns long videos (podcasts, webinars, livestreams, interviews) into vertical 9:16 clips for TikTok, Instagram Reels and YouTube Shorts.

The pipeline, stage by stage

Transcription with word-level timestamps

faster-whisper transcribes the source audio and returns a timestamp for every single word, not just every sentence. Word-level timing is what lets both the clip boundaries and the burned-in subtitles land on the right frame.

Scene cut detection

PySceneDetect scans the video for cuts and hard visual transitions. These boundaries stop a clip from starting mid-shot, which is the most common reason an auto-made clip reads as machine-made.

Highlight scoring

Google Gemini 3.1 Flash-Lite receives the timestamped transcript and the scene boundaries together, and returns the 3 to 15 strongest segments of 15 to 60 seconds. Each one is scored on hook strength, emotional payload and how well it stands on its own without the surrounding context.

Smart vertical reframe

Each clip is cropped from 16:9 to 9:16 in one of two modes. TRACK mode follows a single subject with MediaPipe face detection and a YOLOv8 fallback, damped by a stabiliser that holds the camera steady inside a safe zone instead of chasing every head movement. GENERAL mode handles group shots and landscapes by keeping the full width over a blurred backdrop.

Captions, hooks and effects

Captions are generated from the word-level transcript and burned in with FFmpeg. An AI-written hook overlay covers the opening seconds. Optional Gemini-generated FFmpeg filter chains handle colour and transitions.

Voice dubbing and publishing

With your own ElevenLabs key, dubbing translates the audio into 30+ languages while preserving the speaker's voice, and the dubbed track is re-transcribed so the captions match the new language. Finished clips publish straight to TikTok, Instagram Reels and YouTube Shorts.

Why does moment scoring need the transcript and scenes together?

A transcript alone finds a good sentence but cannot tell whether the shot cuts halfway through it. Scene boundaries alone find clean cuts with nothing worth saying between them. Passing both to the model at once lets it pick a segment that is both quotable and visually intact — the difference between a clip someone would actually watch and one that merely starts and stops in the right places.

Why does the camera hold still instead of tracking the face exactly?

Because a crop that tracks a face frame by frame produces visible swinging, and swinging reads as amateur even when the framing is technically correct. Reframing keeps a safe zone around the subject and only moves the crop when they leave it, then damps the movement on the way. A speaker tracker sits on top to stop the crop flipping between people every time someone nods, and to hold position through brief occlusions.

How long does it take?

On a typical CPU, an 8-minute source video takes roughly 5 to 8 minutes end to end. On an NVIDIA GPU the same video takes about 50 seconds. The gap is almost entirely transcription and encoding; the model call is a small fraction of it.

What does it cost to run?

Clipping AI Cloud puts the GPU and the API keys on us: the first video clipped whole up to 60 minutes, then 20 minutes a month with a watermark and no card, then $12/month for 100 minutes, $29 for 300 or $59 for 750.

Questions people ask

Which AI model does Clipping AI use to find viral moments?
Google Gemini 3.1 Flash-Lite. It receives the word-level transcript with timestamps alongside PySceneDetect scene boundaries, and returns 3 to 15 segments of 15 to 60 seconds scored on hook strength, emotional payload and whether the segment stands alone without surrounding context.
How does the automatic vertical cropping work?
Two modes. TRACK follows a single subject using MediaPipe face detection with a YOLOv8 fallback, stabilised so the crop holds still inside a safe zone instead of chasing every movement. GENERAL handles group shots and landscapes by preserving the full width over a blurred backdrop.
Can it dub clips into other languages?
Yes, with your own ElevenLabs API key, into more than 30 languages, preserving the original speaker's voice characteristics. The dubbed audio is then re-transcribed so the burned-in subtitles match the new language, not the original.

More guides