No abstract

No abstract

Virtual analog modeling of guitar equipment, especially tube-based guitar amplifiers, is a hot topic in sound effect research. However, there are no established standards on the evaluation of the accuracy of such modeling processes. This work presents a first approach on finding a metric for evaluating the similarity between the output of an analog reference amplifier and a digitized version of the same device. A gray-box modeling procedure is used where only assumptions about the general structure of a guitar amplifier are made. The necessary information about the reference device is obtained by input and output measurements. First a digital model is constructed and the error between digital model and analog system is minimized by the Levenberg-Marquardt parameter optimization algorithm. The error is expressed as the result of a cost function. To be able to produce good results, this cost function needs to consider psycho-acoustic aspects like e.g. the perceived frequency resolution. The results of a spectrogram based cost function are compared to the results of a listening test, where the test subjects should rate the similarity of the two audio signals (reference and digital simulation) and the findings are discussed.

No abstract