Text-to-speech Plug-in for windows

Using the text-to-speech plug-in, HomeVisionXL can speak messages generated by your HomeVision schedule. The controller can also pass along the name of a file that should be read aloud.

The plug-in uses version 5.1 of Microsoft's Text-To-Speech system, which is part of windows XP. Users of other windows versions may download the 68 MByte SpeechSDK51.exe from the Microsoft web site. Alternatively, there's a 9.92 MByte download at the MeggieSoft web site that may also work.

Windows XP only comes with the "Microsoft Sam" voice. The SpeechSDK provides the "Microsoft Mike" and "Microsoft Mary" voices. Windows XP users can download these additional voices from the MeggieSoft web site or the Alive Text to Speech site.

This Text-to-Speech plug-in will only work on various Microsoft Windows platforms, which I don't use myself. So the current version of the plug-in is just an example of the possibilities. I hope someone using the plug-in will adopt it, iron out the flaws and come up with new features. Of course I am always willing to help the foster-parent in any way I can.

Speech can be activated by sending a serial string to the main serial port of the HomeVision controller that looks like:

   Serial port 1: Transmit string 'tts string "Hello World"'
Or (for reading out the contents of a file):
   Serial port 1: Transmit string 'tts file "story.txt"'
There are also a number of commands available for controlling the speech:
   Serial port 1: Transmit string 'tts voice "Microsoft Sam"'
   Serial port 1: Transmit string 'tts output "Creative Sound Blaster PCI"'
   Serial port 1: Transmit string 'tts volume "100"'

Text-to-Speech in the Control plug-in

The Text-to-Speech plug-in also creates a new command in the control plug-in. This new command is called "SpeechCommand" and can be used in the actions connected to a control plug-in button. The command takes two arguments. The first argument must be one of: string, file, voice, output, or volume. The second argument is the string to be passed to the specified speech command.

Last modified: 7 October 2006, 11:48 CEST