We are all eagerly awaiting the release of the popular legal siri port i4Siri so I thought I give you guys some information on how the project is going. First of all there they have been working day and they say progress is moving along quite nicely. However, there is no ETA.
If you haven’t read are most recent article on i4Siri then you must know that they have decided to use Google voice speech api. This will process the speech packets sent by an iPhone. Google will process the packets and send the message back to the server. From there the server will process the response with a plugins then send back to the device. To make Siri function properly, they are writing many “Siri Proxy Plugins”. Each plugin would have to handle weather, phone, notes, reminders, iPod, etc. The Siri port will still look and act the same, just the data will be sent and received a little bit different.
The whole article can be read below.
Hi this is Arjun Bhatnagar and I am developer for i4Siri. Everyone has been asking for an update on the i4Siri project and trust me we have been working day and night to provide you with some real updates. There are no ETA’s for the i4Siri release yet, but we are moving along quite quickly.
As you know i4Siri is a Siri port in which Siri connects to Google’s public servers instead of Apple’s. This allows for stability because Apple can block any connection to its servers at any time (which is too risky for a large public server). The connection to Google is only to interpret what you speak. After Google has interpreted your spoken words it is sent to our server. This is where the magic happens.
Currently we have Mark DiFranco working fully on the Google part. He has designed the connection to and from Google and is almost done with our complete break off from Apple’s servers. The past few days he has worked day and night on this and has helped push i4Siri forward. Thanks to his tremendous help we have been able to focus on getting the server ready to make Siri’s functions happen.
To make Siri function properly, we would have to write many “Siri Proxy Plugins”. Each plugin would have to handle weather, phone, notes, reminders, iPod, etc. These plugins are being developed by Alex, Andrew, and I. Development began slowly because the whole concept of Siri Plugins was new to everyone. It took quite awhile for everyone to study and research how the plugins are built. I began and finished Siri’s plugins to understand Web Search, Yelp, and WolframAlpha (the plugin that tells you how many calories in a donut and what 4 times 4 is). When I started work on the phone and calendar plugins I reached a complete stop. There was no method out there on how to connect to these applications. I spent a good day just thinking of how to get these plugins working. After googling around I had an epiphany! I knew how all of Siri’s functions would be programmed and how we could install many custom plugins such as Skype, Facebook, Twitter, etc.
My method to solve the problem was having a Siri website and database. Within our website we have each user who wants to connect to the server register for free and securely store his UDID information. This would first allow us to keep track of who is connected to our server prevent random iDevices from connecting. This means that we can start having information on our users that we can fetch. In order to solve our plugins problem I devised a solution in which we use Google calendar and reminders, online addressbook, online weather websites, etc. So if you wanted to make an event we would first ask you to submit information within your secure account on our website which we then use to connect to Google’s online calendar and post your event. That event would then sync to your iDevice through the calendar’s app and show up there. If you wanted to check the weather we would connect to an online weather database with the zip code you securely submitted and return you with data. This type of solution for Siri’s functions can be implemented easily, it will just take some time.
We are working on i4Siri everyday so please bear with us as we finish the product. When we finish, donators will be able to go to the server url and register an account with their PayPal transaction id.
To test out Siri’s Voice Translation With Google: http://slides.html5rocks.com/#speech-input
If you found this post useful, hit the +1 (recommend) button ?