According to several reports, Apple in the fall will release the full Siri voice recognition agent technology in iOS 6. Current third-generation iPads support voice dictation but not the full Siri artifical-intelligence system introduced in the iPhone 4S.
In a 9to5 Mac article, Mark Gurman reports that the iPad version of Siri will support all functions except phone dialing capabilities.
With the iPad being a device that is larger than the iPhone, the product typically sits farther away from the user’s mouth. Because of this, some speculators of Siri for iPad have believed that the device’s microphone would be unable to properly hear, and therefore incorrectly interpret, command inputs. Sources, however, dismissed this as nonsense, and these sources say that commands input with same or similar accuracy to the iPhone 4S. iPad voice dictation appears to work well for third-generation iPad users.
Gurman says that the iPad 2 and iPad 3 have the processing power and hardware base to handle the voice agent technology. However, Apple is concerned about scaling the server-side capabilities and providing adequate performance to customers, so the Siri for iPad may be confined to a forthcoming fourth-generation model.
Access to Siri by third-party applications is also on the minds of developers in the weeks before the Apple Worldwide Developers Conference (WWDC). On the Tea for iPhone blog, developer Samuel Iglesias, cofounder of Made by Sail, discusses what a Siri API might look like. He says that obviously an API already exists which is used by Apple's apps.
Iglesias considers that the agent programming can be divided into Services and Semantics. Services are the means to accept tasks from another app, something very familiar to programmers (and OSes). The Semantics side is the tricky part, he writes.
The second half of Siri integration, Semantics, is the tricky part: something that most iOS developers have never dealt with. Semantics will attempt to capture the various ways a user can ask for something, and, more importantly, the ways Siri, in turn, can ask for more information should that be required. This means that developers will need to imagine and provide “hints” about the numerous ways a user can ask for something. Sure, machine learning can cover some of that, but at this early stage Siri will need human supervision to work seamlessly.
Iglesias expects that Apple will spend iOS 6 sessions teaching developers the basics of natural language processing and how Xcode will support Siri integration. Or maybe it will be something new.
Let’s keep in mind that a conversational semantic services API of this kind —whatever it will end up looking like—has never been done before and will likely require new tools, new paradigms, to fully capture its power and breadth. And I can’t think of a better place, or time, to introduce such a platform than at the upcoming WWDC.
Very interesting reading. Check it out.