Browsing Centre for Digital Music (C4DM) by Author "Sturm, B"

Now showing items 1-3 of 3

    • BEHAVIOR OF GREEDY SPARSE REPRESENTATION ALGORITHMS ON NESTED SUPPORTS 

      Mailhe, B; Sturm, B; Plumbley, MD; IEEE (2013)

    • Ensemble Models for Spoofing Detection in Automatic Speaker Verification 

      Chettri, B; Stoller, D; Morfi, V; Martinez Ramirez, M; Benetos, E; Sturm, B; 20th Annual Conference of the International Speech Communication Association (INTERSPEECH 2019) (International Speech Communication Association (ISCA), 2019-09-15)
      Detecting spoofing attempts of automatic speaker verification (ASV) systems is challenging, especially when using only one modelling approach. For robustness, we use both deep neural networks and traditional machine learning ...

    • Explaining Predictions of Machine Listening Systems 

      MISHRA, S; Sturm, B; Dixon, S; DMRN+11: Digital Music Research Network One-day Workshop 2016 (Centre for Digital Music, Queen Mary University of London, 2016-12-20)
      We adapt local, interpretable and model-agnostic explanations [1] for use with a machine listening system, and demonstrate it for singing voice detection. Such explanations provide ways to understand the behaviour of machine ...