Vibe Coding DiscoverDeep AI search for every photo and every frame of video in any folder on macOS
A local-first macOS app that searches photos and video using vision embeddings, scene segmentation, OCR, and Whisper transcripts. Optional local LLM chat answers questions about extracted media with citations.
Use Cases
Search photos by natural-language descriptionsFind specific scenes within videosLocate visible text in images and video framesFind spoken words in video dialogueAsk questions about extracted media content
Built With
- Language
- JavaScript
- Frameworks
- Electron · Transformers.js · ONNX Runtime · llama.cpp · Tesseract · Whisper
Tags
local-first · semantic-search · image-search · video-search · computer-vision · embeddings · OCR · transcription · macOS · offline