Snip!t from collection of Alan Dix

<h2>Siri Technology</h2> <div id="heading"> </div> <div id="contents"> <table class="mainTable"> <tbody><tr> <td><p> </p><div id="smallHeading">A New Interaction Paradigm</div> <br> <p>Virtual Personal Assistants (VPAs) represent the next generation interaction paradigm for the Internet. In today's paradigm, we follow links on portals, feeds, and search results. With a VPA, we interact by having a <em>conversation</em>. We tell the assistant what we want to do, in our own words, and it applies multiple services and information sources to help accomplish our task. If the assistant does not understand completely, it asks for clarification. If it cannot find a solution from available sources, it offers alternatives. Like a real assistant, a VPA is <em>personal</em>: it uses information about the individual's preferences and history of interaction to help solve their specific tasks, and it gets better with experience. </p> <p>&nbsp;</p> <p>Siri is the first mainstream consumer application of a Virtual Personal Assistant. It is focusing on the Mobile Internet, where an intelligent interface can help people get things done that would be cumbersome or impractical using the link-following paradigm. For example, in a single spoken utterance the user can ask the assistant to find a nearby restaurant satisfying personal preferences and make table reservations online for a given time. In the traditional paradigm this task would require traversing many links and pages -- and accessing multiple accounts -- making it impractical in many mobile situations.</p> <p>&nbsp;</p> <div id="smallHeading"><strong>Technology for Virtual Personal Assistants</strong></div> <p>&nbsp;</p> <p>The core technology for VPAs corresponds to the essential qualities of an assistant: conversational interface, brokering to multiple services, and personal context awareness.</p> <p>&nbsp;</p> <p><strong>Conversational Interface Technology</strong> - an assistant needs technology to understand speech, natural language, conversational context, and domains of discourse. Siri incorporates state of the art speech to text technology, a patented architecture for recognizing intent in the context of conversation and domain models, and an innovative UI that presents the conversational metaphor in a hybrid of familiar chat-style and search-results interactions that is optimized for small form factors and limited bandwidth environments.</p> <p>&nbsp;</p> <p><strong>Service Delegation </strong>- an assistant brokers the user's intent to multiple services and information sources, gathers and integrates results, and presents them back to the user as a set of solutions to their request. Siri's patented service delegation technology allows it to access many services on the web, combine the results semantically, and return a unified result to the user. In a restaurant selection task, for instance, Siri combines information from many different sources (local business directories, geospatial databases, restaurant guides, restaurant review sources, online reservation services, and the user's own favorites) to show a set of candidates that meet the intent expressed in the user's natural language request.</p> <p>&nbsp;</p> <p><strong>Personal context awareness </strong>- unlike search engines, a virtual assistant will give different answers depending on the individual preferences and personal context (place, time, history). Siri takes maximum advantage of context awareness technology in the modern smart phone combined with cloud-based personalized state management and persistence. As a result, the user gets the advantages of both big computation in the cloud and small computation in the field. </p></td></tr></tbody></table></div>