News of the Institute of Media Informatics

PhD exposition/Haas, G. : "Exploring the Potential of Computer-mediated Auditory Interaction"

Ulm University

Introduction of a dissertation project | Tuesday, 19th January 2020, 14:00 | Video conference

Gabriel Haas, member of the research group human-computer interaction, discusses his phd thesis project.


Abstract:

In the past, visual output devices and haptic input devices dominated human-computer interaction. Thus, much research and development has been done in the field of graphical user interfaces and plenty input devices have been developed for the use by hands. Technological advances in various fields, such as machine learning, semiconductor manufacturing, and signal processing, enabled voice-based interfaces that allow for more natural interaction between humans and technology, and created new ways of processing and interacting with environmental sounds which will be studied and discussed in my dissertation.

In this work, the entities which humans interact with are split into three areas: human-human interaction, interaction with the physical surroundings, and human-machine interaction. In human-human interaction only computer-supported verbal communication was investigated, since the computer as a mediator enables new qualities such as persistence. Asynchronous voice messages as a novel and emerging form of communication were found to be perceived as enabling more efficient communication, help to handle secondary occupations, and better facilitate the maintenance of close relationships as text-based communication in a first work. Building on this, a concept for the manipulation and augmentation of such voice messages was developed and an effective implementation presented.  

As an example of interaction with the physical surroundings, the alteration of the acoustic perception via
wearable audio technology has found widespread application and diverse use. Current headphone usage, as well as demands and applications for future devices, were gathered via focus groups. Based on the insights, the concept of a per audio-source alteration of the acoustic surrounding was developed and tested. It was found to allow for a comfortable, informative soundscape while also satisfying the need for isolation in our dynamic and busy world.

In the setting of human-machine interaction, voice interfaces will be investigated in a final step. The focus of this work will be on increasing naturalness, which is to be realized by incorporating human traits.

Overall, the three identified settings in auditory interaction have been and are being investigated from a theoretical and practical point of view to optimize auditory interaction and present novel possibilities enabled by recent technology.
 

If you are interested in joining the video conference, please contact Mrs Duschek.