Welcome! I'm Daniel Stoller

Research Scientist at Spotify • Berlin, Germany

I conduct machine learning research with a focus on audio and music processing tasks.

Bio

Daniel Stoller is a Research Scientist in the Music Intelligence team at Spotify. He previously received his PhD involving machine learning for music information retrieval from Queen Mary University of London.

His research involves machine learning methods applicable across domains, including generative modelling using diffusion models and generative adversarial networks. He applies these methods to audio and music processing tasks, such as music generation, source separation, lyrics alignment, and multimodal music understanding.

Recent Publications

Efficient Representation Learning for Music Via Likelihood Factorisation of a Variational Autoencoder

Ningzhi Wang, Daniel Stoller, Simon Dixon

IEEE International Workshop on Machine Learning for Signal Processing (MLSP) • 2025

LLark: A Multimodal Instruction-Following Language Model for Music

Joshua P. Gardner, Simon Durand, Daniel Stoller, Rachel M. Bittner

International Conference on Machine Learning (ICML) • 2024

Contrastive Learning-Based Audio to Lyrics Alignment for Multiple Languages

Simon Durand, Daniel Stoller, Sebastian Ewert

IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) • 2023