Michel Olvera (Multispeech) will defend his thesis on Thursday, 15th December at 2pm in room A008.
His presentation will be in English and is entitled “Robust sound event detection”.
Jury members
Thesis directors:
Emmanuel Vincent, Inria
Gilles Gasso, INSA Rouen Normandie
Reviewers :
Mathieu Lagrange, CNRS, LS2N
Juan Pablo Bello, New York University
Examiners :
Anne Boyer, Université de Lorraine
Daniel P. W. Ellis, Google
Abstract:
From industry to general interest applications, computational analysis of sound scenes and events allows us to interpret the continuous flow of everyday sounds. One of the main degradations encountered when moving from lab conditions to the real world is due to the fact that sound scenes are not composed of isolated events but of multiple simultaneous events. Differences between training and test conditions also often arise due to extrinsic factors such as the choice of recording hardware and microphone positions, as well as intrinsic factors of sound events, such as their frequency of occurrence, duration and variability. In this thesis, we investigate problems of practical interest for audio analysis tasks to achieve robustness in real scenarios. Firstly, we explore the separation of ambient sounds in a practical scenario in which multiple short duration sound events with fast varying spectral characteristics (i.e., foreground sounds) occur simultaneously with background stationary sounds. Secondly, we investigate how to improve the robustness of audio analysis systems under mismatched training and test conditions. We explore two distinct tasks: acoustic scene classification with mismatched recording devices and training of sound event detection systems with synthetic and real data.
