Using A Speech-Driven, Anthropomorphic Agent in the Interface of a WWW Educational Application Maria Virvou, Nikitas Sgouros, Maria Moundridou, Dimitrios Manargias Department of Informatics, University of Piraeus, 80, Karaoli and Dimitriou St., Piraeus 185 34, Greece mvirvou@unipi.gr, sgouros@unipi.gr, mariam@unipi.gr Introduction Web-based education has the obvious benefit of allowing platform independent access to easily updated teaching material. Recently, a large number of educational applications have been delivered through the World Wide Web. Unfortunately, most of these systems are static, they do not maintain a student model and thus they are unable to provide individualised tutoring. Nonetheless, the future of web-based education seems promising since researchers in the field of Intelligent Tutoring Systems have made quite successful attempts to either move existing ITSs to the WWW or build from scratch web-based ITSs (Eliot et al., 1997; Stern et al., 1997). However, the development of an ITS requires the involvement of a large number of people, including experts of the specific domain, instructors and programmers. A way to overcome these problems may be the development of authoring tools, which will help construct cost-effective and reusable ITSs in various domains. One such authoring tool for ITSs is WEAR (WEb-based authoring tool for Algebra Related domains). WEAR incorporates knowledge about the construction of exercises and a mechanism for student error diagnosis that is applicable to many domains that can be “described” by algebraic equations (Virvou & Moundridou, 1999). WEAR uses a talking head in its interface with the students. Animated characters have often been used in the interfaces of systems (Rist et al., 1997; Stone & Lester, 1996). Such an interface makes a system more appealing and attractive to the user and if talking about an educational application it may also promote the learning objectives. Walker et al. (1994) investigated subjects’ responses to a synthesised talking head displayed on a computer screen in the context of a questionnaire study. Their findings showed that compared to subjects who answered questions presented via text display on a screen, subjects who answered the same questions spoken by a talking head spent more time, made fewer mistakes, and wrote more comments. In this paper we describe how the talking head is used in the whole educational application. Description of the System and its Interface WEAR aims to be useful to teachers and students of domains that make use of algebraic equations. Such domains could be chemistry, economics, physics etc. In particular the tool accepts input from a human instructor about a specific equation-related domain (e.g. physics). This input consists of knowledge about variables, units of measure, formulae and their relation. When the human instructor wishes to create exercises s/he is guided by the system through a step by step procedure. At each step of this procedure the instructor specifies values for some parameters needed to construct an exercise. Such parameters could be for example what is given and what is asked in the exercise to be constructed. After the completion of this procedure the tool constructs the full problem text and provides consistency checks that help the instructor verify its completeness and correctness. WEAR assigns to each student a level of knowledge according to his/her past performance in solving problems with the tool. The tool suggests each student to try the problems corresponding to his/her level of knowledge. When a student attempts to solve an exercise the system provides an environment where the student gives the solution step by step. The system compares the student’s solution to its own. The system’s solution is generated by the domain knowledge about algebraic equations and about the specific domain in which the exercise belongs (e.g. economics). While the student is in the process of solving the exercise the system monitors his/her actions. If the student makes a mistake, the diagnostic component of the system will attempt to diagnose the cause of it. When interacting with the students WEAR responds through a talking head, representing in some cases the instructor and in some others a co-student. The talking head component of the system uses speech synthesis to automatically produce speech output from text using MBROLA, a freely available speech synthesiser (http://tcts.fpms.ac.be/synthesis/mbrola.html). The talking head renders the interface quite attractive to students through the sound of speech. Moreover, since WEAR is an authoring tool for ITSs