Multimodal natural language platform supporting cellular phones

Abstract

As 3 rd Generation (3G) networks emerge they provide not only higher data transmission rates, but also the ability to transmit both voice and low latency data simultaneously. This capability can be leveraged to provide a multimodal user interface. We describe the end-to-end architecture of our implementation of a multimodal application (voice and graphical user interface) that uses Natural Language Understanding in the speech interface combined with a WAP browser to perform mobile office functions on a cellular phone. A novel aspect of the multimodal platform is that no software is required to be installed on the mobile device. The feasibility of our approach is demonstrated by a successful trial with 50 users over a 3G mobile network. We outline our framework, present the results and observations made during the trial.