A modular and hybrid connectionist system for speaker identification
Neural Computation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Psychoacoustics: Facts and Models
Psychoacoustics: Facts and Models
Musical instrument recognition using cepstral coefficients and temporal features
ICASSP '00 Proceedings of the Acoustics, Speech, and Signal Processing, 2000. on IEEE International Conference - Volume 02
Hi-index | 0.00 |
This paper proposes a classification system based on a hierarchical modular architecture. Such an architecture is designed to address high-dimensional classification problems, that are characterized by a high number of input features and a large number of classes. The hierarchical approach decomposes a "many-class" problem into a hierarchy of "few-class" sub-problems, while the {modular approach} allows efficient treatment of high-dimensional input spaces. Specifically, each classifier in the hierarchy has a modular structure. Each module is implemented as a neuro-fuzzy network that operates on a distinct region of the input space. As an application, the proposed architecture is employed to solve the problem of classifying musical instruments from the sound they produce.