|
|
The SpeechStudio XML
grammar editor uses intelligent navigation and wizards to simplify
and improve the authoring of complex speech-recognition grammars. SpeechStudio XML
grammars improve the power of SAPI 5 XML grammars with extensions for
dynamic grammar creation, in-line semantic actions, and modularity.
SpeechStudio XML grammars are created in manageable pieces that can be
independently evaluated, activated, and deactivated.
Dynamic grammars are automatically evaluated
at run-time, either under application control or automatically as the state of
the application and its environment changes. For example, a grammar based
on a Windows form will be reevaluated when focus changes to the form.
On the other hand, reevaluation of a dynamic grammar associated with a database
can be triggered by the application program.
SpeechStudio grammars feature in-line semantic actions allowing
interactions between the speech-recognition engine and the application to be
specified in the same file. In-line semantic actions have access to each
element of the recognized text, including specific values of dynamic fields, as
well as the ability to control the application, start and stop grammars, and
send information to the application with callbacks.
The Grammar Editor, in combination with SpeechPlayer,
implements a wide variety of infrastructure features allowing rapid and
error-free development of speech-enabled interfaces for telephony, multi-modal,
or Windows applications, such as:
- What Can You Say prompts are
generated automatically from the currently active grammars, and are changed
dynamically to reflect changing grammars.
- Spoken utterances can be recorded in
real-time. The application has immediate access to the recording,
which can then be archived or played back.
- Text-to-speech, pre-recorded
prompts. and audio can be intermixed. They can be synchronized and interrupted.
- Actions and interactions are extensively
logged, with multiple levels of recording and reporting, including application
logging.
- Dictation, confirmation, and speech-engine
configuration are supported.
- Over 100 predefined grammars for integers,
ordinals, real-numbers, times, dates, durations, letters, and spelling.
Features for Multi-Modal Windows
The Grammar Editor investigates the GUI of an
existing program as a starting point for the voice-interface development. From
this starting point, the Grammar Editor helps you to describe conversations and
their program interactions in a familiar Visual Studio environment, associating
voice commands with application menus and dialogs, which are visible in a window
that mimics the Visual Studio Resource Editor.
The Grammar Editor allows your specific
voice commands to control any set of Windows menus, dialogs or controls,
including ActiveX controls. A customized voice user interface can be
created without programming using the extensive library of SpeechStudio
built-in actions. The open interface allows new actions to be created or
custom voice-interface programming to be embedded in the target
application independent of GUI components.
Key Features of the SpeechStudio Grammar Editor
- Applies to new or existing applications without
recoding.
- Familiar look-and-feel of the Microsoft Visual
Studio workspace.
- Creates a customized voice user interface as
value-added for your application, controlled by you.
- Investigates the existing GUI to provide a
starting point for the VUI.
- Wizards produce prototype grammars for menus,
dialogs, and controls.
- Powerful user interface presents command layouts
and their program interactions.
- Open systems design allows users to control voice
interface at finer levels of detail if needed.
- Context-sensitive menus promote quick, error-free
interface construction.
- Extensive tutorial and context-sensitive
documentation provide help when you need it.
- Your GUI will not be compromised by requiring
non-standard "voice-activated" controls.
- Allows application and GUI to evolve
independently of voice user interface.
- Reconciliation wizard synchronizes the voice
interface as your GUI changes.
- Allows voice commands to span multiple activities
or GUI elements rather than requiring control-specific voice commands.
- Integrates user-defined and user-refined speech
elements at runtime.
- Open interface allows integration of third-party
speech objects and controls.
- High-level interface frees users from
non-standard and evolving grammatical notations and the intricacies of
various voice recognition systems.
- Direct integration with your application allows
the best possible voice command articulation with minimum ongoing
investments.
|
