Interspeech'2005 - Eurospeech

Lisbon, Portugal
September 4-8, 2005

Selection of Features and Combination of Classifiers Using a Fuzzy Approach for Acoustic Event Classification

Andrey Temko, Dusan Macho, Climent Nadeu

Universitat Politècnica de Catalunya, Barcelona, Spain

In this paper, we aim to improve the classification of human non-speech sounds produced in a meeting-room environment by using concepts and tools from the fuzzy theory. Starting with an SVM-based baseline system, firstly a reduction of the number of features with the fuzzy measure is shown. And, secondly, a noticeable improvement of the classification performance is reported by combining the outputs of two classification systems with the fuzzy integral.

Full Paper

Bibliographic reference.  Temko, Andrey / Macho, Dusan / Nadeu, Climent (2005): "Selection of features and combination of classifiers using a fuzzy approach for acoustic event classification", In INTERSPEECH-2005, 2989-2992.