Computer Graphics Laboratory ETH Zurich


Scalable Music: Automatic Music Retargeting and Synthesis

S. Wenner, J.C. Bazin, A. Sorkine-Hornung, C. Kim, M. Gross

Proceedings of Eurographics (Girona, Spain,, May 6-10, 2013), Computer Graphics Forum, vol. 32, no. 2, pp. 345-354


In this paper we propose a method for dynamic rescaling of music, inspired by recent works on image retargeting, video reshuffling and character animation in the computer graphics community. Given the desired target length of a piece of music and optional additional constraints such as position and importance of certain parts, we build on concepts from seam carving, video textures and motion graphs and extend them to allow for a global optimization of jumps in an audio signal. Based on an automatic feature extraction and spectral clustering for segmentation, we employ length-constrained least-costly path search via dynamic programming to synthesize a novel piece of music that best fulfills all desired constraints, with imperceptible transitions between reshuffled parts. We show various applications of music retargeting such as part removal, decreasing or increasing music duration, and in particular consistent joint video and audio editing.

Example of our structure-aware music retargeting in the context of video editing.


We applied our algorithm to several applications such as music retargeting (rescaling, singing part removal, structural reshuffling) and video editing (action level control, synchronization). We tested it on a wide range of instrumental musical genres from heavy metal to traditional flamenco. Some representative results are shown in the below video.


Download Paper
Download Video
Download Paper