Automatic singing voice recognition employing neural networks and rough sets

The aim of the research study presented in this paper is the automatic singing voice recognition. For this purpose a database containing singers' sample recordings has been constructed and parameters are extracted from recorded voices of trained and untrained singers of various voice types. Parameters, which are especially designed for the analysis of the singing voice are described and their physical interpretation is given. Decision systems based on artificial neutral networks and rough sets are used for automatic voice type/voice quality classification. Results obtained in the automatic classification performed by both decision systems are then compared and conclusions are derived.