Retrieval Augmented Generation (RAG) is a technique to increase the relevance and transparency of LLM responses. RAG works by supplying the LLM with documents that are relevant to a query. In this tutorial, we'll learn how to perform RAG on audio data using LangChain and Python. APIs: - AssemblyAI API key: https://www.assemblyai.com/dashboard/signup - OpenAI API key: https://openai.com/blog/openai-api Resources: - Code: https://github.com/AssemblyAI-Examples/rag-langchain-audio-data - Blog: https://www.assemblyai.com/blog/retrieval-augmented-generation-audio-langchain/ - LangChain webinar series: null Tools used: - AssemblyAI: https://www.assemblyai.com/ - LangChain: https://www.langchain.com/ - Hugging Face: https://huggingface.co/ - Chroma: https://www.trychroma.com/ - OpenAI: https://openai.com/ ▬▬▬▬▬▬▬▬▬▬▬▬ CONNECT ▬▬▬▬▬▬▬▬▬▬▬▬ 🖥️ Website: https://www.assemblyai.com/?utm_source=youtube&utm_medium=referral&utm_campaign=yt_ry_1 📈 Newsletter: https://www.assemblyai.com/subscribe 🐦 Twitter: https://twitter.com/AssemblyAI 🦾 Discord: https://discord.gg/Cd8MyVJAXd ▶️ Subscribe: null 🔥 We're hiring! Check our open roles: https://www.assemblyai.com/careers ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ #MachineLearning #deeplearning 0:00 Introduction 0:34 Repo and Blog 0:46 What is Retrieval Augmented Generation (RAG)? 0:59 Implementation overview 1:42 Environment setup 2:54 Imports 4:10 Specifying audio files 4:35 Audio document loader 5:12 Text splitting 6:04 Setting metadata 6:29 Text embedding 6:52 Building the Chroma database 7:10 Building the QA LangChain 7:50 Make application loop 8:50 Running the application 9:28 Asking a question and RAG response