Premise New Text-To-Speech driver.

• One or more PCs running Windows XP or Vista.
• Premise Server installed on at least one PC.
• DebugView to observe SYSTester's operation.

• Upon startup, SYSTester reads a configuration file containing the Premise Server's Name and Administrative account.
• Click the Start button to connect to the Premise Server. Upon connection, it installs a new class on the Premise Server, called TTS, and creates a TTS device in Devices > CustomDevices (i.e. see "TTS_DEV004" in the attached image).
• Click the Test Connection button and it will report the connection's status (i.e. it will say either OK or Not OK).
• Click the Stop button to terminate the connection to the Premise Server.

• The target PC can be any networked PC running Windows XP or Vista.
• Ensure the target PC is equipped with .NET Framework 3.5.
• If the target PC is a Premise Server, ensure the MS-SAPI Speech driver is not installed.

1. Create a new folder on the target PC.
2. Copy the attached zip archive to the new folder.
3. Extract the contents.

1. The host name of your Premise Server (do not use "localhost")
2. Username (administrative-level account)
3. Password

1. Open Control Panel and double-click System.
2. Click the Computer Name tab.
3. Use the information found in Full Computer Name.

• Click Start > Run. Type firewall.cpl and press Enter.
• Click the Exceptions tab.
• Click Add Port.
• For Name, enter: SSDP-UDP-3976
• For Port Number, enter: 3976
• Click UDP.
• Click OK.
• Click OK to exit Windows Firewall.

• Click Start > Run. Type firewall.cpl and press Enter.
• Click the Exceptions tab.
• Click Add Program.
• Click Browse.
• Select SYSTester.exe.
• Click OK.
• Click OK to exit Windows Firewall.

' Content: a text string or a file path.
' ContentType: Text, TextFile, XML, XMLFile.
' Voice: one of the listed voices.
' Speak: speak the supplied Content.

with Devices.CustomDevices.TTS_DEV004
	.Content = "<volume level='75'/><emph>Hello world!</emph> I can speak on any PC."
	.ContentType = "XML"
	.Volume = 0.75
	.Voice = .Microsoft_Sam
	.Speak = true
end with

• Supports versioning.
• Better error-handling (i.e. if the RemoteSpeech object is flagged as Disabled or Read-Only).
• If you delete a RemoteSpeech object, its associated Watchdog timer will remove itself quietly after it timesout.

• Now called TTS instead of RemoteSpeechClient.
• Control Volume Level (without having to use an XML tag).
• Set the Audio Format (default is 22kHz 16-bit mono).
• Direct the generated speech to a WAV file (instead of an audio card).

• SYSTester uses Premise's Minibroker.dll to connect to a Premise Server.
• Minibroker.dll is built using 32-bit ATL components.
• SYSTester is compiled as a 32-bit application.
• SYSTester's NXCOMPAT bit is set to NO in order to get around the issue of Data Execution Protection (DEP) and Minibroker's use of pre-version 7.1 ATL components (see Microsoft's knowledgebase).

etc6849 said:

.. It is unclear how to uninstall MS-SAPI as it doesn't show up under "Turn Windows Features On or Off." I could find the associated dll and delete it, but I thought I'd ask before I did this...

Under Control Panel\Speech Recognition Options I have stopped speach recognition is this sufficient?..

Click to expand...

etc6849 said:

..UPDATE: One small issue discovered so far: windows allows hostnames to contain a "-", but for some reason, the Builders IDE doesn't like having a "-" in a devices name when you are scripting....

Click to expand...

		/// <summary>
		/// Return a string that conforms to the naming convention for VBScript variables.
		/// </summary>
		/// <param name="Name"></param>
		/// <returns></returns>
		private string LegalVBName(string Name)
		{
			Regex re1 = new Regex("[^a-z0-9_]", RegexOptions.IgnoreCase); // All non-alphanumeric characters except underscores
			Regex re2 = new Regex("^[0-9].");	   // Leading numbers
			Regex re3 = new Regex("__*");		   // Two or more underscores
			Regex re4 = new Regex("^_*");		   // Leading underscores
			Regex re5 = new Regex("_*$");		   // Trailing underscores

			string result = re1.Replace(Name, "_"); // Replace non-alphanumeric characters with underscores
			result = re2.Replace(result, "");	   // Remove leading numbers
			result = re3.Replace(result, "_");	  // Replace two or more underscores with a single underscore
			result = re4.Replace(result, "");	   // Remove leading underscores
			result = re5.Replace(result, "");	   // Remove trailing underscores
			return result;
		}

1. Ability to adjust the spoken volume (without having to use an XML tag)
2. Set the audio format
3. Direct the generated speech to a file as opposed to an audio card

• Phrase to be spoken can be a text string, XML string, or a file containing text or XML.
• Select the voice (from a list of available voices; list can be refreshed).
• Select the audio output (from a list of available audio cards; list can be refreshed).
• Speak asynchronously (i.e. non-blocking; spoken phrases are in a queue).
• Stop speaking (i.e. purge the speech queue).
• Report driver status.
• Log an event if the driver loses connection with Premise Server.