Clinical RAG System

advanced medical assistant application that utilizes Retrieval-Augmented Generation (RAG) with the Falcon Large Language Model (LLM) to provide accurate and context-aware medical information. Features Audio Interaction Endpoint Speech-to-Text (S2T) conversion LLM processing using Falcon Text-to-Speech (T2S) conversion for audible responses Text-based Interaction Endpoint Direct text input LLM processing using Falcon Text output Retrieval-Augmented Generation (RAG) Enhances responses with relevant medical knowledge Improves accuracy and context-awareness of the AI uses a Retrieval-Augmented Generation (RAG) architecture: User input (text or transcribed audio) is processed. Relevant medical information is retrieved from the Qdrant vector database. The Falcon LLM generates a response based on the user query and retrieved information. The response is returned as text or converted to speech.