MPEG-7-based Low-Level Descriptor Effectiveness in the Automatic Musical Sound Classification

Szczuko P, Dalka P, Dabrowski M, Kostek B. MPEG-7-based Low-Level Descriptor Effectiveness in the Automatic Musical Sound Classification. In: AES Convention 116. Audio Engineering Society; 2004. Paper 6105. Available from: https://aes.org/publications/elibrary-page/?id=12728

@inproceedings{Szczuko2004_12728,
  author = {Szczuko, Piotr and Dalka, Piotr and Dabrowski, Marcin and Kostek, Bozena},
  title = {{MPEG-7-based Low-Level Descriptor Effectiveness in the Automatic Musical Sound Classification}},
  booktitle = {AES Convention 116},
  note = {Paper 6105},
  year = {2004},
  month = may,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=12728}
}

TY  - CPAPER
TI  - MPEG-7-based Low-Level Descriptor Effectiveness in the Automatic Musical Sound Classification
AU  - Szczuko, Piotr
AU  - Dalka, Piotr
AU  - Dabrowski, Marcin
AU  - Kostek, Bozena
T2  - AES Convention 116
M1  - Paper 6105
PY  - 2004
DA  - 2004/05/06
UR  - https://aes.org/publications/elibrary-page/?id=12728
PB  - Audio Engineering Society
LA  - en
AB  - The objective of this paper is to determine which of the MPEG-7 standard low-level sound descriptors are the most significant in the process of automatic classification of musical instrument sounds. First, pitch detection is performed. Then, the parametrization stage of musical sounds based on descriptors contained in the MPEG-7 standard is carried out. Next, a thorough statistical analysis of the feature vectors obtained is performed. For the purpose of automatic classification, two decision systems based on artificial neural networks (ANNs) and rough sets, are used. Both decision systems are trained with feature vectors consisted mostly of parameters contained in the MPEG-7 standard, however their content being reduced after statistical analyses. In addition, a comparison of results obtained by these decision systems with the results got from the nearest neighbor algorithm is made.
ER  -