Re: [Cairo-user] using other acoustic models in Cairo/Zanzibar

[johny j. <joh...@gm...>]

Hello!

Thank you for your valuable answer, SpencerLord. I try my best to
follow all the things which I mentioned and to take into account your
advices.

May I ask you for a favor, please? Let's have a look here:

==================
http://johnyjj2.page.tl/
==================

I know your time is precious so in order not to waste it, I created
something like little tutorial screen by screen about what I have
already configured. It should take much less time than reading long
post. It contains short explanations and some of my doubts. I hope you
can answer those questions.

Mainly they are still about including my own acoustic model in
Cairo/Zanzibar. I know it can be simply done by adding the proper
information to sphinx-config.xml. But if you use WSJ model by default,
I need to create my model very similar in structure to WSJ model.

The second of main issues is the place for specifying the algorithm.
It looks like doing it in similar way to Parrot Speech Application
(http://www.spokentech.org/openivr/writing-speechlets.html) would be
best choice. However my doubts would be more clear after you will have
a look at the site with pictures (http://johnyjj2.page.tl/).

There are also some minor difficulties, they are explained at the top
of the site and in the pictures with red color.

Thanks for help,
regards!