AI-Powered Music Recommendations Based on Mood and Vibe
Qdrant - Vector Database & Search Engine via YouTube
Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Learn how to build an AI-powered song recommendation system in this 28-minute technical talk from Vector Space Talks. Explore how Large Language Models (LLMs) can understand and match the vibe and mood of songs through vector embeddings. Follow along as Filip Makraduli, a Biomedical Data Scientist at Marks and Spencer, demonstrates the creation of a dataset using LLM-generated song descriptions, encoding song vibes with transformer models, and implementing similarity-based vector search to find matching music recommendations. Gain practical insights into combining natural language processing, vector databases, and recommendation systems while discovering how AI can help bridge the gap between musical preferences and emotional states. Access the complete project repository on GitHub to implement your own AI-driven music recommendation tool.
Syllabus
When music just doesn't match our vibe, can AI help? - Filip Makraduli | Vector Space Talks #003
Taught by
Qdrant - Vector Database & Search Engine