Hardware Reference
In-Depth Information
else if ($cmd == "lightsoff") {
system("heyu turn bedroom_light off");
}
?>
You can then work on abstracting and extending this at will. In Chapter 7 you'll integrate this into a
general-purpose message system.
N
Before investing heavily in voice recognition software, ensure that it can distinguish between whatever
different voices can control the system, because a lot of software can listen only to a single, preselected voice as its
primary purpose is dictation and not voice recognition.
Note
Note that most software of this type doesn't provide access to the words you've actually spoken; the computer just
thinks that there's a higher probability of it being
this
one than
that
one. Although this gives you fewer opportunities
for error, it also prevents the use of any analog or scaled commands, such as “dim to 72%.”
From iOS
Something that has partially solved the problem of voice recognition is the cloud. By moving the processing into a
remote device means that you can perform recognition by using more powerful algorithms than would traditionally
be available on a hand-held machine, albeit at the expense of requiring yourself to be online, and finding yourself at
the mercy of a (traditionally) nonopen technology. The most famous example of this has been Siri.
Released in 2010 as a standalone app, Siri has since been incorporated into the iPhone 4s and has been a staple
feature of versions since iOS 5 on both iPhone and iPad. In its traditional mode of operation what happens is:
u
Siri records your voice from the microphone
guzzoni.apple.com
u
Your iOS device uploads a rendering of it to their servers at
u
The guzzoni server processes the voice, and works out the request
u
The guzzoni server returns a command packet to your iOS device
As you can see, two parts of the chain require an iOS device. However, with tinkering (and software such as
SiriProxy) you can circumvent half of this process. The operation then becomes:
u
Siri records your voice from the microphone
u
Your iOS device uploads a rendering of it to your server
guzzoni.apple.com
u
Your server passes the request to
u
The guzzoni server processes the voice, and works out the request
u
The guzzoni server return a command packet to your server
u
Your server can then process any arbitrary command, using the text string that was recognized
u
Your server passes a result string (with optional image) to the iOS device
As you can see, there is a fairly traditional proxy metaphor taking place here, with your server acting as a go
between for the iOS device the guzzoni server so that it is able to step into the middle of the process, to perform
custom commands.
Search WWH ::
Custom Search