Feluda
Open engine for analyzing multilingual, multimodal content at scale.
Made by TattleData and analytics
About Feluda
Feluda is a configurable engine for analysing multi-lingual and multi-modal content. Feluda was built to analyse data collected from social media, including images, text, and video.
The README cites Khoj, a reverse image search engine for finding fact-check articles, as an example use. The repository is licensed under GPL-3.0.
Drafted from github.com · not yet confirmed by Tattle
How it works
Feluda is Tattle's open source engine for making sense of large volumes of mixed media, built for the Indian misinformation and abuse context. It takes text, images, audio and video, runs configurable operators (embeddings, hashing, OCR, transcription, language ID), stores the results, and lets you search and cluster for near duplicates and related content across languages and formats.
It's infrastructure for a fact checking or research team that needs to see how one piece of content spreads in many forms. You deploy it yourself and write the operators you need.
Reviews
Be the first to review it
AnonymousKnow a team weighing Feluda? Send them this page.