How to Convert Long Videos into Shorts with AI

Converting a one-hour podcast or interview into a week's worth of social content used to mean a full day of editing. With PowerCut AI, the same job takes under 15 minutes. This guide walks you through the exact process — from raw upload to published clip.

Step 1: Upload Your Video

Head to PowerCut AI and drag your video file into the upload area, or paste a YouTube or podcast URL directly.

Supported formats include MP4, MOV, MKV, and WebM — virtually everything you'd export from a camera, screen recorder, or video conferencing tool. Files up to 10 GB are accepted.

Once uploaded, PowerCut AI queues the file for transcription and analysis. You'll stay on the processing screen as the next steps run automatically.

Step 2: AI Detects the Best Moments

While you wait, PowerCut AI's pipeline is working through three stages:

1. Transcription — Whisper ASR converts your audio to a word-level transcript with speaker diarisation, so the model knows who's talking and when.
2. Moment scoring — A GPT-class model reads the transcript and scores each segment on shareability, narrative completeness, and emotional intensity.
3. Clip extraction — The top-scoring segments are packaged as candidate clips, each trimmed to start and end at natural sentence boundaries.

The whole process takes roughly 3–5 minutes for a 60-minute video. You'll receive a notification when clips are ready.

Step 3: Review and Select Clips

The clip review screen shows every AI-detected candidate in a scrollable grid. Each card displays:

• A preview thumbnail with the clip duration
• The transcript excerpt for that segment
• A shareability score from the AI (0–100)

Watch any clip inline by clicking its preview. If the boundaries feel slightly off, open the trim editor to adjust in and out points frame by frame.

Select the clips you want to take forward by ticking their checkboxes, then click Proceed to Captions.

Step 4: Customize Captions

PowerCut AI generates word-level captions automatically from the same Whisper transcript. On the caption editor you can:

• Choose a caption style from 17+ animated presets — from minimal subtitle bars to bold highlight-word animations
• Adjust font, size, and position to match your brand
• Edit individual words to fix any transcription errors before export

Caption edits apply to all selected clips at once, so you can set your style once and batch-export everything consistently.

Step 5: Export and Publish

Click Export to render your clips in 9:16 vertical format at 1080×1920, ready for:

• YouTube Shorts — upload directly or schedule via YouTube Studio
• TikTok — supported via direct upload or Creator Portal
• Instagram Reels — upload through the Instagram app or Meta Business Suite

Each clip downloads as an individual MP4 file with captions burned in. If you need a version without burned captions — for platforms that prefer on-platform text overlays — toggle Export without captions before rendering.

Your clips are stored in the PowerCut AI dashboard for 30 days, so you can re-export or re-style them at any time without re-uploading the source video.

That's the complete workflow: upload, detect, review, caption, export. Five steps. Under 15 minutes for most long-form videos. The more consistently you run this process, the faster it gets — each session builds intuition for which clip styles and caption presets work best for your specific audience.