How Siri Works

Voice Recognition on a SmartPhone

By Michael Lagapa

Understanding How Siri Works

Smartphone Voice If you have an iPhone you know that it comes locked and loaded with your own personal sidekick Siri. Siri is the personal virtual assistant that has come included on Apple iPhones 4S and later models. Siri is there to automate tasks and provide information, and Siri becomes more familiar with your tendencies the more you use it. It can inform you of the weather, remind you of an appointment, open an application, and even save your fingers from fatigue by replying to all of your text messages. As you go on asking Siri about the traffic clogging up your commute or what eateries might be lurking about nearby, one question that might come to mind might be; "How can Siri do all of this?"

Brief History of Siri

Apple did not entirely develop Siri by themselves. It started from an AI initiative in 2003 funded by DARPA (Defense Advanced Research Projects Agency) and run by a Stanford University affiliate, SRI International. Their goal was to make a program that helps military personal with office work and making decisions, resulting in CALO (Cognitive Assistant that Learns and Organizes) that learns from its users and the vast amounts of available data. It was used for organizing and scheduling meetings as well as providing the necessary documents for the participants. CALO even made decisions in situations when someone important was unable to make the meeting, they can cancel the meeting and reschedule or continue the meeting as planned if their presence is not needed. Vanguard was another project developed by SRI International that was a prototype that worked very well for smartphones, but did not have the capabilities of CALO.

A startup company that had alumni employees from NASA and Google was formed to combine both the Vanguard and CALO. This company was named Siri and the application they created was called the Siri Assistant. This version would take questions from users via voice or keystrokes, send that data to a remote server to translate, then search various websites. Siri Assistant had a unique kind of humor and a penchant for potty mouth responses. When Apple acquired the Siri company, it stripped down some of its features such as its humor and bad language, and the access to competing websites to prioritize its own services, but it gained multilingual capabilities, iPhone specific features, and gave it the voice we are familiar with today.

Siri's Software: What Can I Help You With?

With the variations and subtle nuances we have in our speech there are an infinite number of ways to construct sentences. Mimicking the complex way humans comprehend speech into a programming software is a monumental undertaking.

Smartphone Voice As a result, Siri’s developers—alongside dictation software company Nuance Communications—have programmed its voice recognition software to interpret commands and questions through a series of steps allowing people to interact with Siri in as human a way as possible.

How Siri Works

Upon receiving your request, Siri records the frequencies and sound waves from your voice and translates them into a code. Siri then breaks down the code to identify particular patterns, phrases, and keywords. This data gets input into an algorithm that sifts through thousands of combinations of sentences to determine what the inputted phrase means. This algorithm is complex enough that it is capable of working around idioms, homophones and other literary expressions to determine the context of a sentence.

Once Siri determines its request, it begins to assess what tasks needs to be carried out, determining whether or not the information needed can be accessed from within the phone’s data banks or from online servers. Siri is then able to craft complete and cohesive sentences relevant to the type of question or command requested.

The Future of Siri

After years of prolific thinking and ceaseless innovation, digital personal assistants are no longer works of science fiction, but are truly amazing feats of technological achievement that get more advanced with every passing year. At this rate, don’t be surprised if holographic personal assistants that you can visualize as well spawn within the next decade. Perhaps the next time you pull Siri up for directions to the grocery store or the occasional philosophical question of the universe, make sure to appreciate all the computing power working behind the scenes to deliver what humans have been doing for years.

Michael Lagapa was a summer intern at Jameco Electronics. He is entering his sophomore year at UC Santa Cruz, and is pursuing a degree in Computer Science. His hobbies include biking, swimming, martial arts, and staying up to date with the latest technology.