I am a PhD student within the ILCC and CSTR institutes of the School of Informatics at the University of Edinburgh. My supervisors are Steve Renals and Peter Bell.

I recently completed an internship with Sonos under the supervision of Daniele Giacobello, and I am part of team Edina in the Amazon Alexa Prize. The data we collected is now available as the Self-dialogue Corpus. I have a small role in the CitySounds project.

My main research interests include

  • speech recognition: factorised adaptation of data in various acoustic and linguistic conditions;
  • computational auditory scene analysis: sound source localisation and separation inspired by the human auditory system; and
  • music information retrieval: inference from audio such as genres, instrument detection, etc., skills which are typically effortless to humans.


I have previously worked as a theatre consultant for Charcoalblue, advising on all aspects of audiovisual and acoustics design; and as a software tester and mechanic for the stage technology company Visual act.



  • J. Fainberg, “Improving Children’s Speech Recognition through Out of Domain Data Augmentation”, Master’s thesis, University of Edinburgh, Edinburgh, 2015
  • J. Fainberg, “Sound Source Separation in Reverberant Environments using Interaural Coherence in a Probabilistic Model of Localisation”, Bachelor’s thesis (Poster), University of Surrey, Guildford, 2014


Chaos and fractals