Navigation Content

Dragon NaturallySpeaking SDK Client Edition

Speech-Enable Your Application


Dragon SDK Client includes everything needed to add cutting-edge speech recognition technology to any Windows-based application. The tools, libraries and ActiveX components of the SDK make it possible to integrate the speech-recognition capabilities of Dragon NaturallySpeaking into any Windows application without using the Dragon NaturallySpeaking user interface.

Download the SDK datasheet for full product details.



Developers can quickly and easily create speech-aware applications “from scratch” or add speech recognition to existing applications. The Dragon NaturallySpeaking speech engine allows end-users to enter, edit and correct text by voice, as well as command and control the application by voice.

Achieve accuracy rates up to 99% Enable continuous dictation into a specified standard Window or any kind of window with more accurate results than the previous release. 

Create New User Profiles Developers can create new user profiles using the Dragon New User Wizard, or from within their applications with or without interaction with the end-user. Version 10 incorporates new acoustic models for better coverage of non-native and regional accents. Together with more powerful adaptation techniques, this will help end-users get better recognition accuracy from the very first use.

The Dragon vocabulary consists of a language model and files containing the words users can say. Developers can create custom vocabularies tailored for any dictation domain.

Developers can leverage the full feature set of the award-winning Dragon NaturallySpeaking, including free text dictation of up to 160 words per minute virtually anywhere you can type; easy correction by voice for complete hands-free operation; and dictation playback for fast editing.

Voice commands Developers can code their applications to respond appropriately when a specified sequence or pattern of words is recognised as a command.

Text to speech Applications can output text in the form of speech by using a voice-text object and the included text-to-speech engine.

Integrate Transcriptions from Audio Files Applications developers can transcribe audio files into application windows or transcribe audio files into text and document files.

Tech Specs

Installation requirements

Each system where you plan to install DSC must meet the following requirements:

  • CPU: minimum 1 GHz Intel® Pentium® or equivalent AMD processor or 1.66 GHz Intel® Atom® processor. We recommend 1.8 GHz Intel Dual Core or equivalent AMD processor. (IMPORTANT: SSE2 instruction set required)
  • Processor Cache: minimum 512 KB. We recommend 2 MB.
  • Free hard disk space: 2.5 GB (2.8 GB for localised non-English versions)
  • Supported Operating Systems:
    Microsoft Windows 7, 32-bit and 64-bit
    Microsoft Windows Vista  SP1 and SP2, 32-bit and 64-bit
    Microsoft Windows XP SP2 and SP3, 32-bit only
    Windows Server 2003 and 2008, SP1, SP2 and R2, 32-bit and 64-bit
  • RAM: minimum 1 GB for Windows XP and Windows Vista, and 2 GB for Windows 7 and Windows Server 2003/2008. We recommend 2 GB RAM for Windows XP and Windows Vista, and 4 GB for Windows 7 and Windows Server 2003/2008 64-bit.
  • Microsoft Internet Explorer(R) 6 or higher (free download at
  • Creative® Labs Sound Blaster® 16 or equivalent sound card supporting 16-bit recording
  • DVD-ROM drive required for installation
  • Nuance-approved noise-cancelling headset microphone (included in purchase). See details at (includes Bluetooth microphones, recorders, and Tablet PCs).


    Note: An internet connection is required for product activation (a quick anonymous process).

Sign Up
Sign up for our newsletter
New in Version 12

See the latest features and enhancements
See What's New in version 12

Contact Us
It is a pleasure to sit in front of my computer talking to Dragon like an old friend. 
-Charles I.
   United Kingdom & Ireland
Client SDK

Choose your country.