In this paper a novel technique for efficient synthesis of waveforms generated by musical instruments is presented. This methodology represents single tones produced by musical instruments as a series of orthogonal Bessel functions, similarly to an additive synthesis that, instead, uses sinusoidal partials. Bessel functions possess a pitch that slowly varies with time, and are thus suited to model musical tones that usually exhibit similar characteristics. A comparative listening test has been performed, and the synthetically created piano sounds have been compared to those generated by traditional additive synthesis. Bessel-based synthesis generally achieved a higher score than the sinusoidal-based approach. The limited amount of memory resources used makes this technique suitable to be implemented on a digital signal processor.
Efficient synthesis of piano tones with damped Bessel functions / Biagetti, Giorgio; Crippa, Paolo; Turchetti, Claudio; Morici, A.. - (2007), pp. 539-542. (Intervento presentato al convegno 2007 15th International Conference on Digital Signal Processing (DSP 2007) tenutosi a Cardiff, Galles, UK nel 1 - 4 Luglio 2007) [10.1109/ICDSP.2007.4288638].
Efficient synthesis of piano tones with damped Bessel functions
BIAGETTI, Giorgio;CRIPPA, Paolo;TURCHETTI, Claudio;
2007-01-01
Abstract
In this paper a novel technique for efficient synthesis of waveforms generated by musical instruments is presented. This methodology represents single tones produced by musical instruments as a series of orthogonal Bessel functions, similarly to an additive synthesis that, instead, uses sinusoidal partials. Bessel functions possess a pitch that slowly varies with time, and are thus suited to model musical tones that usually exhibit similar characteristics. A comparative listening test has been performed, and the synthetically created piano sounds have been compared to those generated by traditional additive synthesis. Bessel-based synthesis generally achieved a higher score than the sinusoidal-based approach. The limited amount of memory resources used makes this technique suitable to be implemented on a digital signal processor.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.