Windows Vista speech engine is part of the Windows Vista operating system and is set to be released sometime in November - December 2006.

Beta testing began earlier this year and news of its "amazing" new built-in voice recognition capabilities spread across the web. As long time Dragon NaturallySpeaking users and solutions providers, we have been waiting for Microsoft to make a big splash.  Let's face it the XP Speech engine was limited in features and most of us considered it to be more of a novelty item than a full featured speech recognition application. 

But if the buzz on-line is correct, Microsoft may wrap up 2006 with a major break through in voice recognition capabilities. For an overview of features, take a look at the latest Vista Speech Video Demonstration created by Chris Henley. 

Based on what we have seen and read thus far, we have compiled a list of important advances. 

1. Seamless dictation, navigation, and application control. The ability to dictate into any application, start and navigate programs with voice commands such as "Start Notepad" and "Switch to Calculator. You can also select, edit and correct dictation by voice. The commands for navigation and correction are very similar to Dragon NaturallySpeaking, making it easy for long time users to transition to the system. To view a list of commands see Rob's Rhapsody.

The beta is a great improvement to the XP speech engine which requires a user to switch between "Dictation" (for dictating) and "Command" (for navigation) Modes. The XP method is unnatural and relatively clunky as users are forced to turn a mode on and off to utilize voice features.  

2. "Document Review" feature reviews documents and sent emails in order to adapt to the users writing style.  It is not clear if this feature allows you to control which documents are analyzed and/or if it also looks for words not already in the vocabulary to enable a user to select what should be included.  

 3. "Say what you see" allows for easy control over navigation by saying what you see on-screen.  In the video demonstration Chris Henley also demonstrates web navigation by voice and utlizes the "Show Numbers" feature to navigate links and images.

There is also support for custom vocabularies and we understand medical and legal vocabularies are already in development through third party entities.  Vista speech is part of the Vista operating system and developers can download a free speech API to easily integrate the technology. Another likely advantage to Vista will be greater stability between the VR application and Microsoft applications (Microsoft Word, Outlook etc.) While Dragon NaturallySpeaking Professional programs are designed to work within these applications, they don't always play nice. For example disabled DLL's that prevent "Select and Say" from working properly from time to time or the strange jumping cursor in Microsoft Word. Less temperamental programs are always a plus and it is safe to expect the Microsoft programs to work more smoothly together.  

With all the momentum building on-line, we are excited and hoping for the ultimate news:   Vista speech supports input from mobile devices!! We love our computers but we also like being away from them, and handheld devices like digital recorders are phenomenal for recording notes, dictation, etc when not connected. We understand it allows for "line-in" as an input option but we are optimistic there will be support for transcription of several file formats i.e. DSS WMA MP3 etc.  For now we wait, research and look forward to More Vista!