Morphing Sound Dynamics

Axel Roebel (1)

1 TU - Technical University of Berlin / Technische Universität Berlin

We investigate a new algorithm for sound morphing that is strictly working in the time domain and is based on a recent method of modeling natural sounds with neural networks called dynamic modeling. By means of basic synthetic signals we investigate the fundamental properties of the proposed morphing scheme. A successful morph of real world sounds of a musical instrument demonstrates the practical application. The results are particularly satisfactory because the time domain algorithm show frequency domain characteristics that are desirable for natural sound morphing applications.

Domaines: Son [cs.SD]; Traitement du signal et de l'image [eess.SP]

Axel Roebel. Morphing Sound Dynamics. 1998 International Conference on Neural Networks and Brain, 1998, Beijing, China. ⟨hal-02911713⟩