Intelligent and perceptual-based approach to musical instruments sound design

作者:

Highlights:

摘要

A novel, generic framework for musical instrument sound design is proposed. The framework uses an intelligent and perceptual based approach to address the two main problems in sound design – optimization of synthesis parameters and assessment of sound quality. A fuzzy model is used to capture and exploit knowledge of sound design from audio experts. A robust methodology, based on the ITU Perceptual Evaluation of Audio Quality (PEAQ) algorithm, is used for objective prediction of sound synthesis quality and sound quality assessment. The new framework makes it possible to automate the optimization of synthesis parameters. It also allows the designer to evaluate objectively, the perceptual impact of individual parameters on the final sound quality which is useful for benchmarking sound synthesis methods. The framework is generic and can be used for sound design for a wide range of musical instruments. We illustrate the use of the framework in pipe organ sound design. Results from this show that the new approach provides an important and useful alternative to existing methods.

论文关键词:Sound design,Audio expertise,Modeling musical instruments

论文评审过程:Available online 29 December 2011.

论文官网地址:https://doi.org/10.1016/j.eswa.2011.12.058