Festival Schedule

Dinosaur vocal calls have been silent since they became extinct following the large asteroid impact event 66 million years ago. Our project, Dinosaur Choir, realizes musical instruments that bring these vocalizations back to life using CT (Computational Tomography) scans, 3D fabrication, and physically-based modeling synthesis. Musicians and gallery visitors give voice to these dinosaur instruments by blowing into a mouthpiece, exciting a computational voice box, and resonating the sound through the recreated dinosaur*s fossilized nasal cavities and skull. When the participant blows into the microphone, their breath becomes the dinosaur*s breath. However, the dinosaur does not have flesh, only bone. They are interacting with and seeing the process of millions of years of decay and change in the instrument. While science is one way of knowing dinosaurs, this work explores how a musical instrument can also generate knowledge. Dinosaur Choir also delves deeper into science, filling the unknowns with informed speculation and imagination.

Our work explores the Corythosaurus, a duck-billed dinosaur with a large, hollow crest housing complicated nasal passages that scientists hypothesize were used for vocal call resonation. Dinosaur Choir begins with this hypothesis but expands its exploration into the unknown vocal boxes, sensorimotor systems, and behaviors that allow vocalization to occur. The project represents a deep collaboration between music, computation, paleontology, and the imagination to explore the intersection of what we know and we may never discover. We have collaborated and consulted with paleontologists to produce this work as well as researching specimens ourselves in university and museum collections. We also explore the dinosaur skull as a wind instrument, iterating on the musical interaction to improve the intimacy and responsiveness of the experience. We focus on breath to drive air pressure into the computational model and we capture the mouth shape of the musician via optical motion capture to determine the stretch and muscle pressure of the vocal folds. This musical interface allows the dinosaur to come alive in the interaction between the musician/gallery visitor and the skull.

This interaction of breath and mouth shape drives the parameters of a the biologically-based bioacoustic syrinx model. Specifically, the amplitude of the participant breath is translated into air pressure below the syrinx membrane (i.e., vocal folds). Generally, the greater the amplitude (and thus, participant air acceleration), the louder the call. The mouth shape of the participant is translated into input muscle tension and pressure in the syrinx model. The wider and more stretched the mouth becomes, generally, the greater the muscle tension in the syrinx, causing a higher pitch. As the mouth area and wideness become smaller, the muscle tension is less and the pitch is generally lower. The physics of vocal mechanisms such as syringes are nonlinear, and both air pressure and muscle tension/pressure affect pitch, timbre, and volume. Participants then, must learn and experience how to use the dinosaur musical instrument in an analogous way that dinosaurs also did.

The computational dinosaur vocal boxes are based on the computational vocal models developed by biologists for birds. We adjust their parameters to reflect estimated and speculated Corythosaurus measurements. The computational nature allows us to implement and allow participants to change which vocal model is being used as well as other parameters. Thus, participants can hear how these different vocal anatomy hypotheses change the sound. We currently have two models implemented based on very different bird vocal anatomies: 1) a raven (Fletcher, 1988) and 2) a dove (Elemans, Zaccarelli, et. al., 2006-9).

Participants will be able to scan a QR code with their phone to use a web interface to change which vocal model and many of the parameters that they are using to create the sound. Thus, they take part in generating and hearing the results of their own hypotheses about vocal anatomy. In this way, they can experience a similar type of informed speculative process that we, the creators, engaged with when creating the instrument, expanding the possibilities. The soft tissue that forms the majority of vocal boxes is rarely preserved, so they also can experience the high scientific uncertainty about dinosaur sound by experiencing the multiplicity of hypotheses that we cannot exclude. Participants also can design sounds according to their own musical concerns as well, going beyond the confines of science into exploring the sounds as pure musical and sounding material.

The CT scans of the adult skull fossil (ROM 1933) were provided by Thomas Dudgeon and David Evans. The fossil specimen (ROM 1933) belongs to the Royal Ontario Museum.