Works fine for me. Normally it doesn't say my name right. So I use phoneme to say it correctly. Here is a snippet:
<speak><phoneme alphabet="ipa" ph="d͡ʒəkwɪntə">Jaquinta</phoneme></speak>
You are using accented characters for this. So you do have to handle them correctly. Make sure your source code is being saved as, say, UTF-8, and that you are compiling it that way, and your editor is set up that way, and that the data stream you are creating uses the correct character encoding. There's lots of places to get it wrong. And it depends on your development environment. But I can confirm that the feature works, as advertised, when you give it the right input.
I really appreciate your help. What you say makes sense. I'm not familiar with working in that mode. When I save my skill in UNICODE format (from WORDPAD), and upload it using the Lambda Management tools, it won't load. Is there a way to get the Lambda Management to upload it correctly, or is it Wordpad. Isn't UNICODE the same as UTF-8 encoding? How do I have control over the compiling if if using the Lambda Management approach?