Master’s Project at Tmh, Kth

This Master’s Thesis investigates the potential benefits and challenges of using a Voice User Interface as an additional access point to a service otherwise only available as a web-based cell phone client using a Graphical User Interface. The service used in the Thesis is a furniture advertisement service. Two different user groups are thought of as the potential end users of this so-called Spoken Dialogue System. The first one consist of people in rural areas in developing countries to whom a Spoken Dialogue System potentially could be the only access point to a web-based service and the second one consists of more advanced cell phone users for whom using a Voice User Interface potentially could have specific advantages compared to using a Graphical User Interface. An end-to-end design life-cycle approach is used as the overall method for this Thesis, including all steps from designing a dialogue structure, conducting user tests to creating a working prototype. The main emphasis is laid upon the two user tests conducted; one conducted in an early stage of the design process using a Wizard of Oz technique and the other one conducted on a working prototype. An open source speech recognition engine called Sphinx4 is used for the final prototype and is also shallowly evaluated. Both tests produce data that, after analysis, generate interesting and useful results regarding weaknesses and strengths of both the specific Spoken Dialogue System developed in this Master’s Thesis as well as general aspects regarding the use of such a system. The most prominent strengths of the investigated system are ease of use, ease of navigation and speed of talking. The most important weaknesses are information overflow, a sub-standard system recognition accuracy and the dialogue being too slow. The two user tests had a transaction success rate of 78% and 90% respectively indicating that the designed dialogue structure and the final Spoken Dialogue System was beneficial even though there were limitations in its implementation. Moreover, both user groups were positive to the idea of using a voice-based interface for the proposed service. Talbaserat gränssnitt som alternativ ingång till webbaserade tjänster En designprocessbaserad undersökning