Sound Research WIKINDX

WIKINDX Resources

Kiss, J., Dahan, K., Soueina, S., & Laliberté, M. 2006, November 22–24, Voice attack recognition for lyric singing learning gesture: An adaptive video game interface. Paper presented at 9th International Conference on Computer Games: Artificial Intelligence and Mobile Systems, Wolverhampton, UK. 
Added by: Mark Grimshaw-Aagaard (1/29/08, 5:11 PM)   Last edited by: Mark Grimshaw-Aagaard (1/29/08, 5:21 PM)
Resource type: Proceedings Article
Peer reviewed
BibTeX citation key: Kiss2006a
Email resource to friend
View all bibliographic details
Categories: Interactive Music
Creators: Dahan, Kiss, Laliberté, Soueina
Publisher: University of Wolverhampton (Wolverhampton, UK)
Collection: 9<sup>th</sup> International Conference on Computer Games: Artificial Intelligence and Mobile Systems
Views: 11/879
Abstract
"We will present an educational video games which propose the user to learn about his own voice attack production in order to teach him how to sing by imitation. The learning of singing whether thought as the construction of the physical vocal pattern, or as the interpretative preparation of a piece of music, proceeds initially through observation, then by imitation of a given physical model. One of the principal difficulties in this mimetic process frequently lies in the different perceptions of the singer and the listener. Without attempting to compensate directly this problem, we propose an interactive system that reacts to external data, thus facilitating the creation of a model of the singer and at the same time real-time simulation of his actions.

We hence offer, within the given framework, to create a basic virtual animation, in order to recognize the voice attack. It is important at this stage to define the nature of the work envisaged, which will not consist of an exhaustive project. We specifically intend to describe the means for encoding the vocal attack expression and its relation to the facial and sound expression, in order to translate a feeling.

We will then proceed to the development of a virtual, computer-generated lyric singer, using the process of interactive phonation as represented by a virtual actyor with simplified expressions. Initially this phonation will be limited to the attack expressions of a small numbher of phonemes. We will also predict the sound development based on this first attack. This part of the process could become the basis of further concrete developments and constitute the foundation for various potential extensions."
Added by: Mark Grimshaw-Aagaard  Last edited by: Mark Grimshaw-Aagaard