|
ENTER THE PUBLIC DOMAIN SPEECHWEB
|
This page contains documentation on how to create SpeechWeb applications of all categories. The documentation gives instructions on how to create
a SpeechWeb application and deploy it on any web server. The page also contains an interactive tool to help developers create simple grammars, a section to download the X+V page, a section containing example grammars and communication protocols and a package to help developers create session type applications.
We have included the documentation for all of our examplar SpeechWeb applications to help developers create similar applications.
A link to a troubleshooting page to solve various kinds of problems encountered while creating SpeechWeb applications is also
available on this page.
For University of Windsor students, we have included a section containing more specific manuals on how to create SpeechWeb applications.
We have documented all of our examplar SpeechWeb applications to help developers create similar applications. The applications are of different categories
and are written in various programming languages. The documentation for each application contains very useful information about the application itself, how was the application created, problems encountered and solutions,
links to the code and a lot more. Below is the documentation for all of our applications, you can also go to the application and say "open your documentation".
Creating a SpeechWeb application isn't complicated. All you need to know is a programming language so you can write a typical text in/text out application. Once you have your application ready,
you would need a few more easy steps to make your application a SpeechWeb application.
SpeechWeb applications can be divided into two categories:
Category 1: user asks questions, computer answers. This type of SpeechWeb applications can be created by following the instructions in the manual Creating SpeechWeb Applications.
Category 2: computer asks multiple choice questions, user answers. This type of SpeechWeb applications can be created by following the instructions in the manual Creating SMS SpeechWeb Applications (Will be available soon). A diagram representing
the architecture of the SMS system can be viewed by clicking here.
For additional help, please go to the troubleshooting page for developers.
To create a SpeechWeb application, you would need to use our X+V page, the use of the X+V page depends on the category of the SpeechWeb application you wish to create, detailed instructions on how to use the X+V page
are available in our manuals. Please click on the button below to download the X+V source code. You need to copy the page into a text file and
rename it as name_of_your_application.xml
The older version of the X+V page can be found by clicking at the link below, this version is still used in many applications such as
Judy. The X+V page was updated in July 2008 to accomodate pictures and to support
session type SpeechWeb applications. The newer version of the X+V can be found by clicking at the link below.
Every SpeechWeb application requires a grammar listing all the possible inputs that are expected to be said by the user. The rule grammar used
by the SpeechWeb architecture is the Java Speech Grammr Format (JSGF). We have developed an interactive tool that lets you create a simple grammar,
please click on the link below to access the tool. If you want to learn how to create a more sophisticated grammr, please go to the documentation of JSGF for detail.
To create a SpeechWeb application of category 2, you would need a system that can manage the sessions for the users using the application at the same time so the program can identify the inputs and outputs.
Therefore, we have developed a package (SMS) that can be used as a session manager for your SpeechWeb application. Instructions on how to use the SMS package are available
in the read me file in the package. A diagram illustrating how SMS works can be found here. To download the SMS package, please click on the button below.
Example grammars written in JSGF and CGI communication protocols are inlcuded in this section:
This section contains detailed manuals on how to create SpeechWeb applications for University of Windsor students.
If your application is written in Miranda,
Please refer to the SpeechWeb Manual - Miranda version.
If your application is written in C programing language, please
refer to the SpeechWeb Manual - C version.