Clips Kitty is an AI-powered video clipping application that converts long recordings into short-form content for social media. It analyzes Twitch, Kick, YouTube, podcasts, and other footage to identify engaging moments using speech, audio, and visual activity. Its multimodal scoring system evaluates potential clips based on excitement, reactions, and narrative impact. Automatic speaker tracking reframes footage into vertical videos suitable for Shorts, Reels, and TikTok. The application generates synchronized captions and provides editing tools for refining clips before export. Users can process content locally with Whisper and Ollama or connect supported cloud AI models. Additional capabilities include multilingual captions, AI dubbing, sports highlight detection, and automated publishing. It is designed for creators who regularly repurpose lengthy recordings.
Features
- Multimodal AI highlight detection combining speech analysis, audio excitement, visual activity, and engagement scoring
- Automatic vertical video reframing with YOLOv8 face tracking and active speaker identification
- Word-synchronized captions with editable transcription, customizable typography, and speaker-specific styling
- Non-destructive video editing with trimming, audio adjustments, transitions, and natural-language AI commands
- Local AI processing through Whisper and Ollama with optional cloud models and hardware GPU acceleration
- Automated short-form video production and publishing with multilingual translation, dubbing, and sports highlight detection