In this step-by-step video, I’ll guide you through a simple web application that harnesses the power of Generative AI. With just a few easy steps, you can upload any YouTube video URL and witness the magic unfold. The application leverages the Assembly AI platform to generate accurate transcriptions of the audio content. But that’s not all! We take it further by utilizing Langchain, a powerful question-answering system, to extract valuable insights from the transcription. The application is powered by OpenAI’s advanced GPT-3 model, enabling you to engage in dynamic and interactive conversations with AI that understand and responds intelligently to your queries. To ensure accurate responses, we incorporate the Chromadb vector store, enhancing the semantic search capabilities of the AI. This means you’ll receive informative and contextually meaningful answers. the user-friendly tutorial will guide you through every step of the process, empowering you to unlock the hidden potential of audio content using the latest AI technologies.
GitHub Link:https://github.com/AIAnytime/Chat-with-Audio-using-LLM
PyTube Gist:https://gist.github.com/AIAnytime/1fb5696a8bfabc27928c4978d4e99272
Assembly AI STT Gist:https://gist.github.com/AIAnytime/14e6affec09fd2de9fedf5eb8c2b1914
High-level Diagram:https://drive.google.com/file/d/1Gm5bCnBUAdrCtIrEQ0Q4A57mFaTf5OFb/view?usp=share_link
ChromaDB Vector Database: https://www.trychroma.com/
Assembly AI Platform:https://www.assemblyai.com/dashboard/signup
Langchain Documentation:https://python.langchain.com/en/latest/index.html
Streamlit Chat:https://github.com/AI-Yash/st-chat
Video Used: https://www.youtube.com/watch?v=W6ZHY0E4_Wg
#generativeai #artificialintelligence #python #coding #chatgpt