I have had this idea in my mind for a couple of years now and would like to get on with it and stop dreaming about it and just jump in and do it already!
I don't mind learning new skills and actually looking forward to getting started, but I need to know that I am not barking up the wrong tree.
I intend to make this an open source project and not as a commercial product.
My dream is to make my phone, my voice-driven computer. Where the phone might not have enough resources, I need to use my main computer as top tier and act as server.
I know ubuntu touch might be a contender but it lacks many drivers for GPS, Propriety drivers, etc...
Let me tell you about my idea first as a high level plan due to my ignorance of the details of this implementation.
I want to make my phone (Nexus 4) be able to listen to my commands via a Bluetooth headset (Through a speech recognition engine), and tell me in in natural speech, things that I need to know or care about. For example: "You have a new email from David. Do you want me to read it or display it?" Sort of like what is now a movie called "Her" without the science fiction and the lovey dovey story part .
I want to write an internet agent that would search the net (from a list of sites) for what I am interested (through an AI program like freehal.net), and keep them for display in the order of relevance when I query it.
I also need to take over the telephony hardware when needed. For example: A call comes in, I want to be able to look-up their number in my database of contacts, if an entry is found, to tell me "Sam is calling. Do you want to answer it?"
I also want to keep a black list table where I know they are trying to sell me something and answer the phone (without me pushing any buttons) and tell them that I am not interested in answering their calls and that they are on my black list. Then hangup. Without sending them to voice mail.
I tried asking the android community on Google 2 years ago to ask why the SDK doesn't provide telephony API front end to answer the call without user interaction, but got no replies.
I've installed ubuntu on my laptop to learn Linux and got it working properly after a couple of days and have downloaded and installed all the programs and sdks I think I'd need. There are a lot of interesting projects in AI and user agents going on and I like to participate in them once I am up to par with the linux environment.
Can anyone give me some guidance, pointers or what to look for. If there are any project in development that I could join, would be much appreciated.
I tried to explain my interest as best as I could. If you need more info, please let me know.
Thanks for all your help guys.