Enriching global apps with Azure Speech SSML

Last reviewed: 9/26/2023

HowTo Article ID: H032313

The information in this article applies to:

  • Chant Developer Workbench 2023
  • VoiceMarkupKit 9

Summary

Microsoft Azure Speech SSML document editing and generation is now available with VoiceMarkupKit.

More Information

Microsoft Azure Speech SSML documents can be created, edited, and tested interactively in the Developer Workbench with VoiceMarkupKit.

Enrich apps with Azure Speech SSML to raise productivity and ease adoption by making them easier to understand and use.

Microsoft Azure Speech SSML expands the capabilities of W3C SSML to harness the power of the Microsoft Cognitive Services Speech Synthesis voices.

There are a few differences with Azure Speech SSML to note as follows:

  • The required voice tag declares the voice used for synthesis. Most other SSML tags are children of the voice tag.
  • The bookmark, lang, and say-as tags extend capabilities unique to Microsoft Cognitive Services Speech Synthesis voices.
  • The speak tag has a new reference to the Microsoft Azure Speech SSML unique tags.
  • The audioduration, backgroundaudio, express-as, silence, and viseme tags provide new unique capabilities.

See the markup reference section Microsoft Azure Speech Synthesis Markup Language (SSML) for descriptions and examples of the Azure Speech SSML tags.

Create, Edit, and Test Azure Speech SSML

VoiceMarkupKit editing surface in Developer Workbench is expanded to support Microsoft Azure SSML syntax. To begin working with Azure Speech SSML, create a document to edit and test.

Create an Azure Speech SSML document

Click the create document button to display the create document dialog, select Azure Speech SSML, and click the OK button.

Create document dialog

The sample SSML document generated in Developer Workbench illustrates how to use Azure Speech with existing W3C SSML tags for a quick start. Easily markup text with the new SSML tags using the intelliprompt assist built into the editor.

Sample Azure Speech SSML document

Leverage all the editing features in Developer Workbench when working with Azure Speech SSML. Explore and hear how new Azure SSML capabilities enabled with tags such as express-as and say-as transform how text is spoken in a variety of languages.

Dynamically Generate Azure Speech SSML

MCSSSMLVoiceMarkup class is available to dynamically generate Microsoft Azure Speech SSML.

VoiceMarkupKit supports markup classes for each markup syntax. These classes provide a common way to dynamically generate markup.

// Generate markup
string markup = _VoiceMarkup.GenerateMarkup("markuptag", "attributes", "text-to-synthesize");

There are common and unique markup tags across SSML syntax. VoiceMarkupKit maps common tags to syntax-unique implementations.

Review the updated reference table in the Generating Markup section to see common and unique tags mapped across SSML syntax.

For each markup tag in the table is one or more links to SSML syntax reference to determine any attribute values if required.

VoiceMarkupKit sample projects in all your favorite development environments are updated to illustrate generating Azure Speech SSML.