I've been using HTML-Kit for around 2 years now. I've not gotten into the depths of the program, but for a free program, it's QUITE capable. The intriguing thing about this program is it's extensibility.

Imagine my surprise when I saw a check box next to "Voice Recognition" in the program. I sat there for a minute wondering what I was seeing. Not much of an explanation on the page, but the help came up with this page and the information below:

http://www.chami.com/html-kit/support/docs/pages/h...

Here are the basics:

HTML-Kit's Text to Speech Wizard is located on the "Tools | Text" menu. Simply select a block of text and invoke the Text to Speech Wizard to have the text read back through the default speech engine. No additional files are required to use this feature on Windows XP and later. Systems running earlier versions of Windows may require the installation of a SAPI 5+ compatible speech engine.

Command Prompt
Experimental support for Voice Recognition (also referred to as Speech Recognition) is provided through HTML-Kit's Command Prompt interface. The default list of command phrases recognized by HTML-Kit is accessible through the "Help | Command Prompt" main menu option. New phrases can be added using the Plug-ins Generator. To enter the Voice Recognition mode, make sure that the "Messages Window" is visible, change to the "Command Prompt" tab on the Messages Window and check the "Voice Recognition" option. If a SAPI 5+ compatible speech recognition engine is installed on the system, HTML-Kit will enter the Voice Recognition mode at which point any of the default or custom phrases can be used to invoke commands associated with them.

So this program can use Text to Speech and will use SAPI 5+ speech engines along with the ability for custom phrases. They have links on the above page to Microsoft but their page is really confusing.

When I get some time I intend to play with this. You can get HTML-Kit here: http://www.chami.com