Efficient modal sound synthesis on GPUs

Modal sound synthesis is a useful method to interactively generate sounds for Virtual Environments. Forces acting on objects excite modes, which then have to be accumulated to generate the output sound. Due to the high audio sampling rate, algorithms using the CPU typically can handle only a few act...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2014 IEEE VR Workshop: Sonic Interaction in Virtual Environments (SIVE) S. 13 - 18
Hauptverfasser: Rausch, Dominik, Hentschel, Bernd, Kuhlen, Torsten
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.03.2014
Schlagworte:
Equations
Force
Graphics processing units
Instruction sets
Kernel
Mathematical model
Vectors
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Modal sound synthesis is a useful method to interactively generate sounds for Virtual Environments. Forces acting on objects excite modes, which then have to be accumulated to generate the output sound. Due to the high audio sampling rate, algorithms using the CPU typically can handle only a few actively sounding objects. Additionally, force excitation should be applied at a high sampling rate. We present different algorithms to compute the synthesized sound using a GPU, and compare them to CPU implementations. The GPU algorithms shows a significantly higher performance, and allows many sounding objects simultaneously.
DOI:10.1109/SIVE.2014.7006284