The Dragon NaturallySpeaking SDK Server Edition is designed for integration into solutions that require speech recognition to run in the background, without any interactive dictation or self editing by the users.

	Optimized Accuracy From Any Dictation Source

• Leverage specialized acoustic models to support input from close talking microphones, hand held digital recorders, or telephones
• Continually increase accuracy rates as misrecognitions are corrected, either by the user or a third party editor
• Includes Nothing But Speech™ umm and ahh filter
• Includes an expanded pre-configured language model library of 20 medical specialties
• Create unlimited custom vocabularies tailored for any lexicon
• Control speed/accuracy ratio to optimize for required performance levels

	Flexible Architecture Simplifies Integration

• Develop with any language that supports ActiveX, including C++, C# and Visual Basic.
• Automate user profile creation with no requirement for interactive script reading.
• Scalable development and run-time solution that allows you to use as many or as few engines as needed by your application.
• XML concordance file allows for easy synchronization between recordings and text output.
• Natural punctuation eliminates the need to dictate periods and commas.

	Powerful Engine Maximizes Processing Power

• The Dragon NaturallySpeaking SDK Server Edition can run 24 hours a day, 7 days a week, and processes recordings in real time or faster.
• Each server can recognize up to 12,000,000 lines per year*.

Distributing Your Application To distribute Dragon NaturallySpeaking SDK technology as part of your application, you must have an OEM licensing agreement. Please Note: Any license to distribute Dragon NaturallySpeaking technology does not include a license to incorporate or distribute RealSpeak™ Text-to-Speech engine. For more information on licensing for both RealSpeak and the Dragon NaturallySpeaking SDK, please contact Synapse Sales.