Discussion:
Tokens ...
(too old to reply)
Raj
16 years ago
Permalink
What is a token? and how a message is splitted into tokens?? and how many
tokens will be created???

Thank you

Regards

Raj
William DePalo [MVP VC++]
16 years ago
Permalink
Post by Raj
What is a token?
A token is a light weight object that contains all the attributes necessary
for the eventual instantiation of the object which it represents. (It's an
implementation detail on which you can't rely but the voice tokens are
actually registry keys).
Post by Raj
and how a message is splitted into tokens?? and how many tokens will be
created???
What message?

Regards,
Will
www.ivrforbeginners.com
Raj
16 years ago
Permalink
Dear Will,

Thanks indeed for the reply and your precious time!

Friend of mine is doing a project using Speech SDK and she like to know how
audio message is splitted into tokens and how into how many tokens?

Example:
The audio text:
"Hello Will, How you are you"

How is the above audio text splitted into tokens and into how many
tokens.

It was very difficult and find and get answer for this question through
search engines

Thank you again

Regards

Raj
Post by William DePalo [MVP VC++]
Post by Raj
What is a token?
A token is a light weight object that contains all the attributes
necessary for the eventual instantiation of the object which it
represents. (It's an implementation detail on which you can't rely but the
voice tokens are actually registry keys).
Post by Raj
and how a message is splitted into tokens?? and how many tokens will be
created???
What message?
Regards,
Will
www.ivrforbeginners.com
William DePalo [MVP VC++]
16 years ago
Permalink
Post by Raj
Thanks indeed for the reply and your precious time!
You are welcome.
Post by Raj
Friend of mine is doing a project using Speech SDK and she like to know
how audio message is splitted into tokens and how into how many tokens?
"Hello Will, How you are you"
How is the above audio text splitted into tokens and into how many
tokens.
It was very difficult and find and get answer for this question through
search engines
SAPI's tokens aren't used for that. I think that your friend has made some
assumptions that are not valid. IMO, if she tell us what she wants to do
maybe someone can help. Her goal is not clear at all. You say you have
"audio text". The first step is to say whether she has audio or text?

Regards,
Will
Raj
16 years ago
Permalink
audio
...
William DePalo [MVP VC++]
16 years ago
Permalink
audio
Ok, then there are two options - dictation or command and control.

In any case, this is only close to the truth (I've not seen the source of
the recognizer).

For dictation, the engine is looking at the incoming digitized audio and
trying to detect bits of words called phonemes. It compares the phonemes
against the phonemes it finds in the audio stream against those that make up
words in its dictionary. When it thinks it has got a match (deciding that's
quite complicated) it returns a guess as to the most likely word and repeats
until there is silence.

For command and control, the task is simplified because the recognizer needs
to compare the audio it hears against only the phrases that it has in the
active grammars.

The nice thing about using SAPI is that your friend needn't be concerned
with the details - SAPI operates at a much higher level. Besides they are
far too complicated for most of us without PhDs in the field and lots of
years of experience.

If your friend has a real need to understand this stuff at that level of
detail she can take a look at this

http://cmusphinx.sourceforge.net/html/cmusphinx.php

which is an open source project out of Carnegie Mellon University where a
lot of research into natural languages is done or this page from MS research

http://research.microsoft.com/en-us/groups/srg/

though I don't know if they publish the source (I doubt it).

Regards,
Will
Raj
16 years ago
Permalink
Thanx will

Very informative!!

Regards

Raj
...
Loading...