Peter Foster

I completed my PhD at the Centre for Digital Music, at Queen Mary University of London; my thesis work examined using information-theoretic measures of predictability for the purpose of determining musical similarity in audio-based music content retrieval. I have subsequently worked as a postdoctoral researcher on identifying sounds in domestic environments, in collaboration with Audio Analytic Ltd; and as a postdoctoral researcher on automatic music transcription, in collaboration with DoReMIR Music Research AB.

Research interests

Contact information

If you would like to contact me, feel free to send an email or connect via LinkedIn.




The CHiME-Home dataset is a collection of annotated domestic environment audio recordings. Labels are assigned at the level of 4-second chunks. The set of possible labels is child speech, adult male speech, adult female speech, video game / TV, percussive sounds, broadband noise, other identifiable sounds.


pyitlib is a library of information-theoretic methods for general data analysis and machine learning tasks, implemented in Python and NumPy.

Reviewing Activities

Journals: Conferences: