Speech Grammar Editor Features

Speech Grammar Editor provides a graphical view of the grammar incorporated into an application, breaking it down into different paths and displaying them visually. Highlighted features of Speech Grammar Editor are outlined in the following section. For a more complete description, see Enabling Speech Recognition in the Speech Application SDK (SASDK) Help documentation.

Highlighted Features

• W3C Grammar Support
  Speech Grammar Editor opens and saves files only in the World Wide Web Consortium (W3C) Speech Recognition Grammar Specification (SRGS) format, which is the grammar format supported by the Speech Add-in for Microsoft Internet Explorer and the Speech Add-in for Microsoft Pocket Internet Explorer.

• Semantic Authoring
  The editing tools support the W3C specification for adding semantic information to grammars. To simplify this process, the SASDK provides two semantic tool windows. One window enables simple assignments for properties and values in one pane, with more detailed variable settings in another pane. The second window provides an easy interface for monitoring semantic properties.

• Semantic Script Editor
  The Semantic Script Editor supports editing the contents of the SRGS tag element, which associates semantic values with words or phrases in the recognition grammar. The Semantic Script Editor contains an assignments page that contains aids for creating a restricted set of scripts and a test pane for unaided script authoring.

• Pronunciation Editor
  The Pronunciation Editor enables developers to look up word pronunciations in a copy of the core lexicon of the speech recognition engine, and combine or edit the phonemes that specify the pronunciations of words. The Pronunciation Editor contains a phoneme palette for easy reference to the phoneme set used in the lexicon. Also included in the Pronunciation Editor is a feature that plays back a pronunciation in text-to-speech (TTS) synthesis to help guide developers as they build or modify pronunciations.

• Features Facilitating Use of Referenced Rules
  A speech input recognition grammar consists of a structured list of rules that identify words or phrases that can be used for speech recognition. Rules describe specific sequences of words or phrases, and are defined by the content of rule elements contained in a grammar. While writing a rule that contains a reference to another rule (defined by a ruleref element), developers can easily open and edit the referenced rule by either double-clicking the ruleref object in the rule, or opening the target rule through a contextual menu. In addition, Speech Grammar Editor exposes comments that immediately precede a rule element in a .grxml file as the description for that rule in the Description field of the Rule Browser, enabling developers to identify at a glance which input each of the many different rules in a large grammar recognizes.