# Local Processing

RLAMA is a local document question-answering tool that connects to a local Ollama model to provide users with document indexing, querying, and interactive functions. It supports multiple document formats, ensuring that data is processed entirely locally to protect privacy and security. This tool is primarily aimed at developers and technical users to improve the efficiency of document management and knowledge retrieval, particularly for handling sensitive documents and private knowledge bases. The current product is an open-source free version, with potential for future feature expansion.

doesVideoContain is a model that leverages artificial intelligence to detect video content directly in the browser. It allows users to automatically capture video screenshots and identify significant moments in videos through simple English sentences. This model operates entirely on the client side, preserving user privacy without the need for API fees, and can handle large local files without uploading them to the cloud. It utilizes Transformers.js and ONNX Runtime Web from the Web AI ecosystem, combining custom logic to perform cosine similarity calculations.

aTrain is an offline speech transcription tool developed by researchers at the Center for Business Analytics & Data Science, Graz University, and tested by researchers at the Graz Knowledge Center. It utilizes the latest machine learning models to automatically transcribe audio recordings without uploading any data. aTrain is featured in a paper published in the *Journal of Behavioral and Experimental Finance*. Please cite this paper if used for research. It supports Windows 10 and 11 systems, with users able to download and install it through the Microsoft App Store or the BANDAS Center website. Installation guides for Linux systems are available on the Wiki. aTrain's primary advantages include privacy protection by avoiding data uploads, high-quality transcription accuracy, and fast processing speeds on your local computer.