The W3C has published the first Working Draft of Natural Language Semantics Markup Language for the Speech Interface Framework, part of the voice browser activity.

The voice browser activity takes as its basis VoiceXML and several other voice/speech markup languages, and is addressing the creation of specifications for speech grammars, voice dialogs, speech synthesis, natural language representation, multimodal systems and reusable dialogue components.

The newly release draft addresses the issue of natural language representation. From the introduction:

It is expected that this markup will be used primarily as a standard data interchange format between Voice Browser components; in particular, it will normally be automatically generated by a semantic interpretation component to represent the semantics of users' utterances and will not be directly authored by developers.

The language is focused on representing the semantic information of a single utterance, as opposed to (possibly identical) information that might have been collected over the course of a dialog.

As an example, the draft gives some markup for the phrase "OK," when it is intended to be interpreted as "yes" by an application.

<result x-model="http://theYesNoModel"
        xmlns:xf="http://www.w3.org/2000/xforms"
        grammar="http://theYesNoGrammar>
  <interpretation>
      <xf:instance>
          <myApp:yes_no>
              <response>yes
          </myApp:yes_no>
       </xf:instance>
       <input>ok
   </interpretation>
</result>

The NL Semantics Markup Language uses the data models from the W3C XForms working draft.