Using Synthesized Audio to Improve Workspace Awareness in Distributed Groupware
Workspace awareness – the up-to-the-moment knowledge of who is in a shared space, where they are, and what they are doing – is an important factor in smooth and natural collaborative work. Awareness is more difficult to maintain in distributed settings than during face-to-face work; to overcome this limitation, several kinds of awareness displays have been designed that provide information about collaborators. Most awareness displays are visual, however, and therefore suffer from three limitations. First, displays must be visible in order to be useful, but many scenarios (e.g., both small-screen and large-screen settings) make it difficult for users to see the awareness information. Second, visual information about activities may be difficult to see if the action is small or if the workspace is cluttered. Third, the observer must attend to the awareness display in order to see it, but as tasks become more demanding, it becomes more difficult to notice changes in the display. Visual presentation, however, is not the only option for awareness information. Audio information is a natural part of shared activity in the real world, and previous research has shown that non-speech audio can successfully be used to help maintain awareness in groupware systems. Audio has several advantages that can overcome the drawbacks of visual awareness displays: audio takes no space and does not need a location on the screen; audio is not affected by workspace clutter; audio can be processed without requiring visual attention; and audio can be used in parallel with visual information. In addition, audio can be effective for several types of awareness information, such as whether activity is occurring right now, when actions start and stop, where actions are happening, the type of activity, and the qualities of the action (e.g., lines drawn slowly or quickly). In this project we are evaluating the use of audio awareness as an enhancement for visual awareness displays. We explore two main questions: can dynamic audio generated with granular synthesis adequately convey information about the type and character of off-screen activities; and can the addition of audio information improve on a visual awareness display, particularly in situations where visual presentations are difficult to see or attend. We have carried out studies to examine the quality of the synthetic sound by asking people to determine the shape, speed, and pressure of off-screen drawings, using synthesized chalk sounds and matched recordings of real chalk, and to compare visual and audio awareness in a real groupware system where participants had to carry out an individual task and also stay aware of off-screen activity. Our results so far show that the addition of audio awareness information significantly improves people’s ability to stay aware of off-screen events, compared to the visual radar view alone.
We have developed a usable toolkit for developing natural awareness sounds using granular synthesis; the toolkit is available at the U of S HCI Lab website.
Images and Videos
Chalk shapes used in the recognition study
- Gutwin, C., Schneider, O., Xiao, R., and Brewster, S. (2011) Chalk Sounds: The Effects of Dynamic Synthesized Audio on Workspace Awareness in Distributed Groupware, Proceedings of ACM CSCW 2011, 85-94.
NSERC – SurfNet