Semantic Markup Language Reference

The following topics in this section describe the structure and content of SML output and serialization of semantic results into Semantic Markup Language (SML) output. A speech recognition engine generates SML output from grammar files that conform to the Speech Recognition Grammar Specification (SRGS) 1.0. For more information about the grammars, see SRGS Grammar XML Reference and Semantic Interpretation for Speech Recognition (SISR) Version 1.0. The SML output in these topics can be duplicated by the use of the ConstructSmlFromSemantics() method, using the code example shown on that page.

In This Section

• SML Output Overview

  Describes the structure and content of SML output.

• Serialization of the Root Rule Variable

  Describes how the recognizer serializes the value of the root Rule Variable.

• Serialization of Developer-Defined Rule Variable Properties

  Describes how the recognizer serializes the value of Rule Variable properties that developers define in semantic interpretation markup.

• Serialization of Multiple Rule Grammars

  Illustrates how the recognizer serializes grammars with multiple rules.