Title: Relative harmonic magnitude extraction for musical instrument detection.
Abstract: Harmonic structure is widely used for assessing the timbre of pitched musical instruments. However, fundamental frequency and overtones are not easy to extract due to spurious partials and missing fundamentals. This study attempts to extract underlying harmonic structure based on source-filter modeling. The spectral envelope is first extracted by linear predictive analysis. To discard spurious partials, the source is divided out using noise level estimation, yielding harmonic structure characteristic to particular instruments. The resulting relative harmonic magnitudes, as well as the spectral envelope, are used to classify musical instruments. Experiments were conducted on the RWC musical instrument sound database. Results show that features are preserved for each musical instrument, over varying pitches.
Publication Year: 2011
Publication Date: 2011-04-01
Language: en
Type: article
Indexed In: ['crossref']
Access and Citation
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot