World Opera Project

The World Opera Project is a distributed, real-time live opera performance planned to take place simultaneously in several Canadian, U.S. and European cities.  At the Shared Reality Lab (SRE), we had the opportunity to run a series of experiments aiming to look into "how audio and video transmission delays affect the emotional connection between singers", "the choice of strategies singers use to compensate for these delays" and "the effect of orchestra placement on the resulting performance".


The set of experiments was a collective effort from all of the members at the SRE lab, in collaboration with Niels Windfeld Lund from the University of Tromsø. My role in this project involved collecting and understanding the singers' views and experiences of singing remotely, analyzing any behavioural changes experienced while singing under various audio and video delay conditions (for instance, we looked at the gaze interactions between singers, and singing latencies).

We found that while the delay conditions often led to undesired changes in tempo, these might be overcome, in part, through rehearsal. One can therefore appreciate the importance of the rehearsal process under actual delay conditions when putting together a full operatic production between remote performers. This would not only help to fine tune the balance between the synchrony and asynchrony of the performers, but would also help the singers develop and practice a coping strategy for the very specific delay introduced by their remote location.

The role of the conductor turned out to be important in maintaining the flow of the musical piece and preventing the singers or the pianist from following each other, which would have resulted in the familiar “recursive drag on tempo”. Nevertheless, we observed that the difficulty in judging how to compensate for the audio and video delay precluded any adjustment by the conductor for musical expression. While the initial trial of the entire experiment was difficult for the conductor, he, along with the pianist, developed a system in which they ignored the audio feed coming from the singers after a few more trials. As the conductor stated: “I do not correct because I can't judge.”

Responses from participants to the questions related to their emotional connection with the remote singer indicate that latencies in the range we were testing had little or no impact on perception of presence, even though these may have been significant to the musical result. Rather, we found that the biggest factor influencing presence was the singer's familiarity with the technology. The longer the experiments proceeded, the less the singers indicated distraction from the camera-monitor mediated experience and a greater sense of emotional connection to the other musicians, regardless of latency. More details about these experiments and findings can be found in the publication listed below.


 Olmos, A., Brulé, M., Bouillot, N., Benovoy, M., Blum, J., Sun, H., Lund, N.W., and Cooperstock, J.R. (2009), " Exploring the role of latency and orchestra placement on the networked performance of a distributed opera ", 12th Annual International Workshop on Presence, Los Angeles, Nov. 11-13, 2009.