This project is supported by the HP Voice Web Initiative in the HP philanthropy program.
In this project we integrate differrent information services, such as our local transportation information service Busman, an e-mail reading service Mailman, and ubiquitous information and messaging service Doorman in a unified service center, which will be available through a common speech interface. Existing services, which are accessible by using both speech and web interfaces, will be integrated to work together. This new service will be targeted at two kinds of users. First, we will set up ubiquitous information stations in an office environment. Services will be available both to the office employees and to the visitors of the office. Secondly, the same services will be available to mobile users through cellular phones and personal digital assistants. The services will be adapted to the situation in hand and new interaction techniques will be implemented to meet the demands of different use situations. Finally, a user study will be performed by equipping a group of users with high-end PDA’s (Jornada 720) and by studying how services really get used in different contexts.
The results of the project are:
First phase
Objectives: development of general software architecture, hardware setup (servers, mobile devices, wireless network), informing users about the project (web pages, user database, advertising).
Description: in the first phase of the project we will design and implement the general architecture for the new service concept. The architecture will be a set of extensions to our general speech application development framework called Jaspis. At the same time we set up the hardware infrastructure (two servers, 30 mobile PDA devices, wireless network) and inform the participants (30 people containing employees from our laboratory, regular visitors to our laboratory and other voluntary users) about the available services.
Second phase
Objectives: integration of the available services (bus timetable information, e-mail reading, visitor guidance, phone call relaying) and new interaction techniques.
Description: in the second phase the new service concept is implemented by integrating existing services. Phone-based services, including Mailman (an e-mail reading service) and Busman (a local bus timetable information service) will be modified to provide information to other applications as well. Furthermore, their existing telephone interface is used as a basis to provide telephone access to other services, too. Our ubiquitous office environment, Doorman, will be extended to support mobile clients (PDA’s with wireless network connections) and telephone users. Finally, we will implement a speech interface for mobile devices to access the different services.
Third phase
Objectives: user study, reports and articles, final version of the applications and the system architecture.
Description: in the third phase we will carry out the user study by equipping the users with mobile devices. We use laboratory tests to study specific interaction issues. This work will be done using our two well-equipped usability laboratories. In addition to laboratory tests, we will perform field tests and analyze the real use of the services by office, mobile phone and PDA clients.
First phase
Infrastructure and users: we have designed, implemented and tested the overall system infrastructure. We have set-up the wireless network access points (AP-1000) and application servers (Omnibook 6000). We have distributed the handheld devices (Jornada 720) to the test users (30 employees from our laboratory) in order to make them familiar with the machine. We have informed the test users how to set-up mobile devices (including wireless network access) and how to use different existing services. This has been done by using e-mail, group portal (www-page) and giving personal guidance. We have also arranged meetings where usage scenarios and ideas for new applications have been discussed. The user database has been collected.
System architecture: we have started the work to extend our Jaspis architecture to support a large number of mobile devices to access different information services. This includes the design and implementation of the middleware architecture, which provides basic services for mobile devices. Main focus areas are speech synthesis, speech recognition and messaging services. Different software and hardware solutions have been tested during this phase, especially speech synthesizers, short message servers and audio equipment. We have also investigated new techniques, such as Bluetooth and GPRS. This work continues in the next phase of the project.
Applications: in order to integrate different speech-based information services, such as our e-mail application Mailman and ubiquitous computing application Doorman in a unified service center, we have made an integration plan which describes how existing services can be made available using mobile devices. Furthermore, we have designed new functionality to the existing applications for the users of mobile devices. This includes the use of location awareness information and accessibility of the mobile users. We have discussed with visually impaired people in order to build some applications, such as a speaking SMS service, for them. We have also been designed how existing non-speech services, such as our group portal can be speech-enabled and made more accessible by using mobile devices.