New interest in spoken dialog research has recently been generated by the advent of SIRI(TM). This application has ignited the imagination of many who have begun to believe that speaking to automatic assistants is possible and useful. In the near future everyone will need to use speech to communicate with technology such as Google Glass and smart watches which can't use a keyboard. Using speech will also give wider access to technology, for example to the elderly and the disabled. It will serve as an economic motivator, spurring new technology and technology-based products. The vision of the Real User Speech Data project and community infrastructure (RUSD) is that it will serve the spoken dialog community by sparking the creation of streams of spoken dialog data from real users, by distributing data and by helping researchers access spoken dialog platforms where they can compare their findings with those of others. It will link high school and undergraduate students (who have lived with the technology all of their lives, and can imagine how they would like to talk to technology), with technology experts who can implement their ideas. RUSD will serve the needs of the spoken dialog community at the same time by helping this community to use these novel and useful speech applications to record and collect speech of their users. RUSD will help to distribute this speech data to the community. It is essential in making successful systems as it will be used to retrain them (adding data makes the statistical representations more precise) and to assess them (comparing systems helps find which novel techniques improve the systems) and to create new ones. In this manner, RUSDs contribution will lead to both higher quality research and to widespread use of speech technology. The current planning project is dedicated to the discussion of the future RUSD with the research community which includes a hands-on Challenge contest to finalize RUSD's functional requirements, and to the preparation of a comprehensive RUSD proposal to the CISE Research Infrastructure Program.
The overarching goal for the RUSD project and community infrastructure is sparking new initiative that will generate data and research platforms for the community. It will relate high school and undergraduate students' ideas for novel, transformational applications of speech technology with the current spoken dialog community's technology. The resulting real applications will provide the streams of data from real users who are interested in repeatedly using an application because it helps them in some way (solving some task, or entertaining them). This will allow researchers to make their systems more robust. Unsolved basic research issues in spoken dialog include: signal processing in noise, recognition of groups of "difficult" users (like the elderly and non-native speakers), management of complex dialogs (i.e. in meetings and with agents), and the automatic use of meta-linguistic information such as prosody. RUSD will help support new applications so that they can be used as research platforms where these and other issues can be explored. The Dialog Research Center (DialRC) has provided this service for one application, Let's Go, which gives bus information in Pittsburgh. Since then research has evolved, necessitating real dialog systems in more challenging areas. One site cannot create and run all of them. RUSD can help the community to create these systems, insuring that they meet the needs of the community and obtaining consensus on a unified infrastructure. The planning phase of RUSD includes: (1) querying the spoken dialog community to define its needs going forward; (2) conducting a Challenge to determine interest in producing real streams of data and platforms; (3) gathering tools that can be used for the Challenge systems and ensure their documentation; (4) preparing the full proposal to the CISE Research Infrastructure Program for RUSD's design and implementation.
