Page 8 - index
P. 8
Voice Control: HEY CORTA- Voice Recognition & Control System. The
client collects input from the user and sends
NA, OK GOOGLE, SIRI & it to the Server where all the really complex
ALEXA computing is accomplished. The Server anal-
By Phil Sorrentino, Contributing Writer, The yses the input and develops the responses
and sends them to the Client where the re-
Computer Club, Florida
sults are presented to the user in audio and/
or display formats.
http://scccomputerclub.org / Phil-
sorr.wordpress.com Naturally Speaking is certainly a useful prod-
philsorr (at) yahoo.com uct, but the voice recognition and control that
has really gotten the attention of the public
Remember Dragon Naturally Speaking? It lately, are the intelligent personal assistants
was, and still is, Voice Recognition software that are provided by some of the leading
mostly used to control the operation of a word computer companies, Apple, Microsoft,
processor like Word. Certain words were used Google, and Amazon. Apple was first on the
for very specific manipulation of the cursor and scene with “Siri”, followed by Google’s “Ok
the text. Naturally Speaking came on the sce- Google”, then, with Windows 10, came Mi-
ne and became useful sometime around 1999 crosoft’s “Hey Cortana”, and finally Amazon’s
to 2003, depending on how much you needed “Alexa”.
to transcribe documents into the computer.
Early versions had to be “trained” by the user All of these are Client-Server implementa-
to recognize their individual voice, and the tions. The Servers are somewhere in the
speed and accuracy were sometimes accepta- cloud and the Client resides on your
ble, and sometimes not so much. Things have smartphone, in the case of Siri and “OK
really improved since then; now the manufac- Google”, or on your laptop (or desktop, or
turer, Nuance, claims in its advertising that tablet) in the case of “Hey Cortana”, or on a
“Dragon is 3x faster than typing and it’s 99% special device that is placed centrally located
accurate”. So, Voice Recognition software has in your home, in the case of “Alexa”.
really come a long way.
All of these assistants use a Natural Lan-
(For those of you, who are not familiar with guage User Interface to answer questions.
Naturally Speaking, it has three primary areas You’ll need a microphone on your device to
of functionality: dictation, text-to-speech, and take advantage of this capability. The Client
command input. The user is able to dictate app, on the device, uses the microphone to
and have their speech transcribed as written listen for a “Wake Phrase”. After this phrase
text, or they can have a document synthesized is recognized, the following intercepted
as an audio stream, or they can issue com- speech is then sent to the Server where it is
mands that are recognized by the program.) analyzed via speech recognition software,
and converted to commands. The Server
Naturally Speaking is an example of a local then uses these commands to gather an-
computer application or App. All the computing swers to the original spoken inquiry. All of
needed for it to operate is on the computer these assistants can make recommendations
that runs it. Naturally Speaking doesn’t take and perform various actions via their Server
advantage of Client-Server technology. If you capabilities. (For example, a verbal request
attended one of our classes, you will recall that for the “weather” might yield various audible
when an application is implemented with Client statements about the weather in your loca-
-Server technology, the heavy lifting (computer tion. Or, a request for “traffic” might yield au-
processing) is not done locally, but rather at a dible indications of the traffic in your location,
Server that is very powerful and very fast, but or possibly maps indicating traffic problems.
remote from the Client. The remote Server is Or, a request for the best restaurant might
connected to the Client by the internet, which yield a list of restaurants near your location.
allows rapid movement of data between the Or, if you have things set up, the statement
Client and the Server. So the Client App runs “Add eggs to my shopping list” will yield an
on the local computer and is connected to the updated shopping list including eggs.)
Server Software, running in the cloud, via the
internet. This combination provides the total
8
