How do I tailor SSML by language and voice?

Last reviewed: 11/7/2023

HowTo Article ID: H032315

The information in this article applies to:

  • Chant Developer Workbench 2023
  • VoiceMarkupKit 9

Summary

Microsoft Azure Speech Synthesis is available in 149 language and 466 voices. Not every feature is supported in every language by every voice. Generating voice-specific SSML ensures the best sounding results.

More Information

Not all Azure Speech SSML is supported by every voice and attribute values vary such as the following:

  • say-as format attribute is based on the value of interpret-as attribute value,
  • express-as role attribute is based on the voice name attribute value, and
  • express-as style attribute is based on the voice name attribute value.

VoiceMarkupkit editor IntelliPrompt offers a more productive way to assist in knowing what options are available and avoid having to remember all the rules for every voice or search the documentation every time.

IntelliPrompt enhances SSML editing productivity with text completion popup lists and popup tips that provide more information about what is being edited.

The IntelliPrompt completion list automatically appears when typing SSML. A completion list can be scrolled and item selected with a mouse click. Alternatively, arrow keys can navigate the list and the tab key select an item.

To force the IntelliPrompt completion list to display, press the ctrl+space keys at the same time when the cursor is in attribute value position.

VoiceMarkupKit Azure Speech SSML editor IntelliPrompt in the Developer Workbench provides popup lists and popup tips for the all the Azure Speech SSML. It also goes further in suggesting the appropriate values based on other previously defined attribute values for:

  • speak xml:lang,
  • voice name,
  • mstts:express-as role,
  • mstts:express-as style, and
  • say-as format.

To display the list of languages voices support, either type <speak xml:lang=" or place the cursor there and press the ctrl+space keys.

Azure SSML language completion list

To navigate down the list, either scroll or type the language letters of interest.

Navigate language completion list

To select a voice available in the language, either type <voice name=" or place the cursor there and press the ctrl+space keys.

Create an Azure Speech SSML document

To select the correct format for time (or date, or duration) interpret-as value, either type <say-as interpret-as="time" format=" or place the cursor there and press the ctrl+space keys.

Create an Azure Speech SSML document

To select an express-as style value, either type <mstts:express-as style=" or place the cursor there and press the ctrl+space keys.

Create an Azure Speech SSML document

To select and express-as role value, either type <mstts:express-as role=" or place the cursor there and press the ctrl+space keys.

Create an Azure Speech SSML document

Note if the IntelliPrompt completion list does not display, there are no values defined currently for the SSML as declared for the current voice and language.