Agilingua's Technology

For successful communication between humans, speech production and perception by the ears are not enough. In addition, contextual knowledge and communication strategies prompting us to reply, ask, suggest or answer are required to ensure effective communication. The software equivalent to these intellectual functionalities of understanding and decision-making on a case-by-case basis is called dialogue management.

Agilingua offers the original spoken dialogue engine- the ATOM Dialogue Engine- contained in a complete speech interface development kit, the ATOM Spoken Dialogue SDK.

Natural Language Understanding + Dialogue Management

Most existing advanced speech interfaces utilize NLU (Natural Language Understanding) technology, which allows the system to understand the user's free-form natural speech. Very often, extra emphasis is placed on its capability to allow the user to speak freely, and the interface is often promoted as �gconversational.�h

However, NLU alone does not make the system speak back to the user �gconversationally.�h All possible speech responses by the system are predetermined and custom programmed, so it merely a mechanical reaction.

In addition to the NLU, Agilingua�fs speech interface employs dialogue management technology, which empowers the system to intelligently take the initiative in creating speech sentences and responding to the user based on multiple factors such as previous interaction and situation.

The artificial intelligence of the dialogue manager dynamically generates the response of a reply, suggestion, question or answer in real-time, resulting in a more human-like conversational interaction.

The dialogue manager plays the compulsory role, especially in more complicated, longer conversations which take multiple turns of interaction between the system and the user.

Agilingua is one of a small number of firms with extensive expertise in dialogue management and natural language understanding. Our dialogue technology evolved from research at Carnegie Mellon University and University of Karlsruhe, two of the world�fs leading research institutions for natural language processing. We offer products and services for the development of natural conversational voice dialogue interfaces for embedded and server-based systems.

Application areas of our dialogue technology include in-car telematics and infotaiment, Interactive Electronic Technical Manuals, robotics and more.

For technical details, please see Development Pages.

�@�@