Bringing together commercial and academic perspectives for the development of intelligent AmI interfaces

Abstract

The users of Ambient Intelligence systems expect an intelligent behavior from their environment, receiving adapted and easily accessible services and functionality. This can only be possible if the communication between the user and the system is carried out through an interface that is simple (i.e. which does not have a steep learning curve), fluid (i.e. the communication takes place rapidly and effectively), and robust (i.e. the system understands the user correctly). Natural language interfaces such as dialog systems combine the previous three requisites, as they are based on a spoken conversation between the user and the system that resembles human communication. The current industrial development of commercial dialog systems deploys robust interfaces in strictly defined application domains. However, commercial systems have not yet adopted the new perspective proposed in the academic settings, which would allow straightforward adaptation of these interfaces to various application domains. This would be highly beneficial for their use in AmI settings as the same interface could be used in varying environments. In this paper, we propose a new approach to bridge the gap between the academic and industrial perspectives in order to develop dialog systems using an academic paradigm while employing the industrial standards, which makes it possible to obtain new generation interfaces without the need for changing the already existing commercial infrastructures. Our proposal has been evaluated with the successful development of a real dialog system that follows our proposed approach to manage dialog and generates code compliant with the industry-wide standard VoiceXML.

An approach to develop intelligent learning environments by means of immersive virtual worlds

“…However, as described in Section 2, virtual worlds and Sloodle have been successfully used by teachers of different levels and with different degrees of technical knowledge. With respect to our proposal, we have worked in rapid prototyping of conversational agents by non-technical users [25], and are currently working to employ our last advancements to make the architecture totally transparent for the teachers that will use, so that they just have to supply a simple model for the conversations expected.…”

“…Speech recognition and synthesis are performed using the Microsoft Speech Application Programming Interface (SAPI), integrated into the Microsoft Windows 7 operating system. We have also integrated modules for semantic understanding and dialogue management [25], which are based on speech grammars following the W3C SRGS specification 10 and Voice Extensible Markup Language (VoiceXML) files, 11 the standard defined by the W3C for speech ac-cess to Internet information and services. VoiceXML files are specified as conversational finite state machines in which users are in a conversational state, or dialogue, at a time.…”

Developing multimodal conversational agents for an enhanced e-learning experience

“…The design of conversational agents has reached a maturity based on standards that pervade technology to provide high interoperability, which makes it possible to divide the market in a vertical structure of technology vendors, platform integrators, application developers, and hosting companies [PIERACCINI, R., et al 2009], [GRIOL, D., et al 2012b] . The implementation and strategies of conversational agents employed in e-learning applications vary widely, reflecting the diverse nature of the evolving speech technologies.…”

“…Conversational agents as personal coaches integrate information about the domain of the application. For example, Grigoriadou et al [GRIGORIADOU, M., et al 2005] describe a system where the learner reads a text about a historical event before stating their position about the significance of an issue and their justification of this opinion. Similarly, in the CALM system [KERLY, A., et al 2008b] the users answer questions on the domain, and state their confidence in their ability to answer correctly.…”

“…Using natural language in educational software allows students to spend their cognitive resources on the learning task, and also develop more socialbased agents [RODRÍGUEZ, S., et al 2011]. For this reason, this kind of agents have been employed to develop a number of educational systems in very different domains, including tutoring [PON-BARRY, H., et al 2006], conversation practice for language learners [FRYER, L., et al 2006], pedagogical agents and learning companions [CAVAZZA, M., et al 2010], dialogs to promote reflection and metacognitive skills [KERLY, A., et al 2008b], or role-playing actors in simulated experiential learning environments [GRIOL, D., et al 2012a], etc. To successfully manage the interaction with users, these agents are usually developed following a modular architecture, which generally includes the following tasks: automatic speech recognition (ASR), spoken language understanding (SLU), dialog management (DM), database management (DB), natural language generation (NLG), and text-to-speech synthesis (TTS).…”