research

Have a look on the right for my recent publications. My research interests include the following topics within music information retrieval and bioacoustics:

  • music recommendation
  • automatic music tagging
  • scaling to large music catalogs
  • computational (ethno)musicology
  • digital humanities
  • biometrics, ecg and ppg signals
  • raspberry pi, arduiono and the internet of things
  • automatic score following
  • social tagging and folksonomies
  • automatic animal sound detection

Have a look at my Google Scholar site for my publications.

Music Recommendation

R&D Data Science / Music Recommendation at Utopia/Musimap

Until recently I worked in a great Data/AI team, which developed and operationalised music recommenders for the music industry.

Audio Defects Detection

Post-doc at IRCAM

We created a method using Deep Convolutional Networks to detect audio defects automatically – e.g. for scanning large catalogues of digital music data prior to distribution to customers. I was able to follow through this exciting project as a post-doctoral researcher at the Analyse-Synthèse group at IRCAM, Paris. The research included both creation of a source of relevant and realistic defects in audio files, as well as development of the detector. Some of the results are presented in this paper at ISMIR.

Dig That Lick

From the Trans-Atlantic Digging into Data Challenge

In collaboration with City University of London, Queen Mary University of London, Columbia University, CNRS, IRCAM, University of Music Franz Liszt, University of Illinois Champaign

With the aim of linking (automatic) transcriptions of Jazz solos, pattern matching and discographies, this project brings together renowned scholars and results from several high-profile projects. We developed an dataset and interface enabling a fresh perspective into jazz history. Have a look: web interface. More info on our project page.

The Digital Music Lab

An AHRC Digital Transformations Project

In collaboration with City University, UCL, QMUL, British Library

During my post-doctoral research at City University I worked on the DML and ASyMMus projects. Both part of the Big Data call of the Digital Transformations in the Arts and Humanities Theme we created a methodology and infrastructure including a web interface and API for remote extraction and access of analysis results from datasets provided by the British Library, I Like Music and the CHARM project.

Spot The Odd Song Out

Similarity Model Adaptation and Analysis using Relative Human Ratings

Finished PhD Studies at City University download thesis

I have recently finished my PhD thesis within the Music Informatics Research Group at City University London. You can read more about the thesis here. A major factor which allowed this project to be successful was our game Spot The Odd Song Out. With Music Technology Group at Universitat Pompeu Fabra, we have developed a flamenco version! You can try the game here. Spot The Odd Song Out

Detecting Bird Sounds via Periodic Structures

A Robust Pattern Recognition Approach to Unsupervised Animal Monitoring

Diploma Thesis; Bonn University 2008 download

My diploma thesis evolved during a collaborative project of the Multimedia Signal Processing Group of Bonn University and the Animal Sound Archive in Berlin. (Long term) monitoring recordings have been performed in a nature conservation area at Parstein Lake, Brandenburg, Germany. Besides a general description of the Project, the thesis focuses on general signal processing algorithms for bird songs featuring periodic structures. These are currently used to identify the Savi’s Warbler calls within such recordings. Furthermore, techniques are provided for the analysis of more complex rhythmic structures. The design of the algorithms heavily orients towards robust recognition of the repetitive song, as the undirected recordings are subject to varying grades of (weather-induced) noise. Key technologies include: FFT, Novelty Curves, Autocorrelation, Hidden Markov Models.

Wuner

A basic windows guitar tuner

Software, Win32 download

This Software may contain errors or malfunction, use with care!