This week I'm getting some online training ready for my current client and discovered that Adobe Captivate and the Paul voice from Neospeech had problems with the following sentence:

"Numbers are also present on signs throughout the building."

At first I couldn’t quite put my finger on what was wrong with the way this was being said.  I listened a few times and realized that the word present sounded like the noun present that you would give on a birthday or at Christmas, rather than adjective that I am present in this situation or the sign is physically present in the building.  Technically the words are both pronounced the same, but there is a stress difference between the two.  Clearly I meant to say that the signs are physically located throughout the building.  I could rewrite the sentence but instead I did a little research and discovered a better solution. 

I wondered if there was a way to let the text to speech engine correct this.  I've discussed before the need to inject pauses using either additional commas.  You can also add pauses with more control using commands such as or .  The 500 in the first example means 500 milliseconds, while Break level 2 is similar to the pause that a comma produces. 

Turns out there is a solution for changing the pronunciation or stress on certain words that change depending on the parts of speech used.  Here is the syntax:

text

In my specific example, I entered:

present

This ends up changing the pronunciation ever so slightly from a present as in a gift, to the state of being present time or physically here right now.