Apple stunned its users with new technology advancement as SIRI. SIRI works by letting you use your voice to send messages, make phone calls, schedule meetings and more. Siri which was first started as DARPA project at SRI International in early 2003. Later they build their own setup as Siri International (SI) in 2008 and launched the Siri Digital Assistant as a free application on the Apple’s App Store. In a Couple of months Apple acquired SI for nearly around $200 millions. Apple then made advancements in its backend architecture and integrated it with iOS.
Siri is a better solution because it process natural language. You no need to learn or improve your accent to say words clearly enough. Siri is far beyond that. It understands language, applys logic and learns it.
Implementation of various technologies to make SIRI works:
- Siri’s artificial intelligence – an engine which is used to process natural language
- Nuance Communications’ voice recognition and Text-to-speech (TTS) technology – it translates the voice into text.
- Backend services which includes data processing abilities, access to data and various other resources
At the iPhone level, Siri records and plays voice files, maintains communication with Apple’s data center and interacts with iPhone apps. Siri Backend handles all the other heavy lifting.
Siri architecture totally relies on backend processing. This is done possible by the mobile processor, memory and storage. Limitation with Siri is that it can’t be used without network access. But Apple’s team believes that now or later the network access will become faster and will be available on every phone. The main advantage of Siri Architecture is that if you want to extend the services it can be done centrally and simultaneously available to all users.
How Siri Works
- In order to use SIRI you should be connected to internet, without internet the SIRI won’t work.
- Open the Siri App by long pressing the Home button on iphone.
- Once it’s opened you should command it to perform particular operation and that command is recorded and compressed and then sent to Apple’s data center.
- Now the backend processing starts, the Nuance speech- to-text and AI-like natural language processing figures out what has been said.
5. The Nuance speech- to-text , translates the voice file into text and performs a particular operation depending upon the request placed by the user by accessing t0 Apple’s data center and services.
6. After the complete processing, the text file is converted to voice file and send back to the phone.
7. Where it is uncompressed and it is played on the handset with response to the query or request.This is how it works.
Apple has integrated Siri with iOS5 and now we are pretty good to use it. Integrating Siri with iOS5 actually involved recording the words or requests, compressing them and sending them to Siri Backend where the voice file is converted to text and processed, maintaining the communication between the iphone and Apple’s data center and sending back the text file which is processed by converting it into voice file, uncompressing and playing Siri response files at the same time on the handset. Everything from meaning understanding, learning, turning speech into text to uncompressing and sending back the processed voice file is done by the Siri Backend.
Siri Technology is not the one of its kind but it got lots more competition like Android’s Voice Actions, Edwin, Iris, Vlingo and many others…