The Apple iPhone 4S was much like many other hot device releases, with the new product featuring slightly enhanced features and specifications, with one notable exception. The iPhone came with what Apple promised would be a revolutionary new technology that changed how we interacted with the iPhone: Siri.
The virtual personal assistant was the result of years of research into how we communicate. But now that the technology has been out for a while, the question remains whether the product lives up to its initial hype.
The Problem Siri Solves
Interacting with smartphones has given people headaches for a few years now. There’s so much promise there, but often it becomes difficult to find exactly what you need, especially as more and more specialized applications take on more specific tasks. The problem winds up being a lack of communication between the person and the device.
Take, for instance, the task of finding a nearby restaurant. First you would have to open the maps program, or another GPS-based application, then manually search for restaurants and scroll through a list to find one you like. If you then want to schedule an appointment to meet someone there, you would have to go back out of the program, into another application, and manually enter the event.
Siri looked to change all of that.
The Basis of the Technology
Siri works by listening to your voice and then performing the task that you ask it to do, which often includes opening other programs and entering or changing information. For example, if you tell Siri “Tell Bob I’m running late,” the program will find Bob in your contacts and send a text message that you’re running late.
Now, Siri isn’t the first voice-based assistant — Android had one at least a year earlier. But the difference between Siri and the earlier voice recognition software is night and day. Siri is based on the Cognitive Assistant that Learns and Organizes (CALO), initially developed by the Pentagon and then later by a special project built specifically for the technology. This new form of voice recognition, which is really a rudimentary form of artificial intelligence, centers around picking apart natural language to understand just what the speaker is saying, and then performing an action or issuing a response based on the request.
Not only can Siri interact with many of the apps on your phone, but since the artificial intelligence is run by powerful servers that the service connects to, it can also interpret and answer a large range of questions. If you are cooking and really need to know how many cups are in a pint, just ask Siri and it’ll instantly give you an answer. If you want to know what tomorrow’s weather will be like, you’re just a quick phrase away from knowing.
And this is where the true genius behind Siri comes out. You don’t have to be very specific when you ask it questions. “What will the weather be like tomorrow?” “Will I need an umbrella tomorrow?” and “Will it rain tomorrow?” will all get you the same answer. The software also remembers earlier requests, so if you ask it to find a hamburger joint in the area, and don’t like what you see, by simply saying “How about tacos?” Siri will understand that you’re still looking for a restaurant.
It’s this shift from requiring the user to speak very specific phrases to just requiring them to speak as if they were talking to a human that separates Siri from the crowd.
While the news surrounding the new technology was extremely positive when the iPhone 4S was first released, over time some people have soured slightly on Siri, and some of these people took to sites like ZD Net and The New York Times to express their frustration. Everyone can pretty much agree that Siri was a big step forward over previous voice assistants, but that doesn’t mean that the technology is perfect.
First, it needs a connection to its servers to function, which means Siri will be useless if you don’t have a signal, even if you just need it to manipulate something in your phone.
Second, tests have shown that Siri isn’t all that great at answering questions, giving the correct answer less than 70 percent of the time. The software also has some trouble accurately interpreting the question in noisy environments like a city sidewalk. The combination of incorrect answers and misconstrued questions can quickly make one wonder if the service is more hindrance than helpful.
The Natural Language Future
Even with these few minor drawbacks, there’s little doubt that Siri has started a revolution in how we think about interacting with our mobile devices. Apple has invested plenty into the service, both in development and marketing, so you can be sure that future versions of Siri will increase both its performance and the number of applications it can interact with.
Other similar services will undoubtedly come out in the future, but the fact is that after Siri, voice recognition that doesn’t understand natural language is an obsolete technology. We want more from our smartphones now, and after experiencing Siri, even the smooth physical navigation on an iPhone or Android device will seem dated and clunky.
Even with its existing flaws, Siri succeeded in showing us the future of device interaction, and as the technology matures you can expect fewer errors, more integration, and more adoption.