MHP Interactive Applications: Combining Visual and Speech User Interaction Modes

This paper proposes an architecture for the development of interactive digital television systems accessible via voice. The proposed architecture is applicable to any interactive digital television application developed under the MHP standard, which runs on an interactive receiver equipped with return channel via Ethernet. The user needs only a mobile device with audio I/O, capable of running a speech synthesis and recognition system which follows the W3C standards SSML and SRGS respectively. This work includes the formalization of the architecture to meet this objective and validates it through two use cases in the scopes of eHealth and Digital homes.