Language Models for Music Recommendation

Explore the innovative application of language models in music recommendation systems in this 32-minute conference talk from EuroPython 2023. Delve into the unique approaches taken by popular streaming services like Spotify and YouTube to personalize content recommendations. Discover how complex recommendation systems incorporate various factors such as track metadata, user information, and play counts alongside the track content itself. Learn to build your own music recommendation system using language models, inspired by Andrej Karpathy's work on GPT. Gain insights into addressing user curiosity about recommendation algorithms and understand the potential of applying natural language processing techniques to enhance music discovery experiences.

Language Models for Music Recommendation — Nischal Harohalli Padmanabha, Raghotham Sripadraj