In this project, we investigate how vocalizations produced with movements can support the design of sonic interactions. We propose a generic system for movement sonification able to learn the relationship between gestures and vocal sounds, with applications in gaming, performing arts, and movement learning.
The system is based on Hidden Markov Regression (HMR) [francoise2013multimodal] to learn the mapping between sequences of motion features and sequences of sound descriptors representing the vocal sounds — for example, MFCCs. During the demonstration phase, the user produces a vocalization synchronously with a gesture. The joint recording of motion and sound features is used to train a multimodal HMM encoding their relationships. For performance, we use HMR to continuously generate the sequences of sound descriptors associated to a new movement sequence, that drive the synthesis of the vocal sounds using descriptor-driven granular synthesis.
For SIGGRAPH'14 Emerging Technologies, we created an imitation game based on the vocalization system [francoise2014mad]. The first step is to record a vocal imitation along with a particular gesture. Once recorded, the systems learns the mapping between the gesture and the vocal sound, allowing users to synthesize the vocal sounds from new movements. In the game, each player can record several vocal and gestural imitations, the goal is then to mimic the other player as precisely as possible to win the game!
Vocalization is an essential component in dance and movement practice, and is often used to support movement expression. Many choreographers use vocalization to communicate to dancers a set of attributes of the movement, such as timing and dynamics. In Laban Movement Analysis, vocalization is used to support the performance of particular Efforts relating to movement qualities.
We conducted a study on the sonification of Laban Effort factors using the vocalization system [francoise2014vocalizing]. We trained a system using expert performances of vocalized movement qualities, that we used in an exploratory workshop to support the pedagogy of Laban Effort Factors with dancers.
Synekine is a project by composer Greg Beller that “brings together performance and scientific research to create new ways to express ourselves. […] In the Synekine project, the performers develop a fusional language involving voice, hand gestures and physical movement. This language is augmented by an interactive environment made of sensors and other Human-Computer Interfaces...”
Greg Beller used both our systems based on Hidden Markov Regression and Gaussian Mixture Regression in his prototypes “Wired Gestures” and “Gesture Scapes”: