|
The idea of speech recognition has been around for as long as the most basic forms of electronics, going back to the Audrey system created by Bell in 1952 and the Shoebox system created by IBM later in 1962. DARPA then pushed the development of speech recognition forward between 1971 and 1976, with the potential for military applications being the driving force behind this research. We're getting away from the topic though - so how does Dragon Medical Practice Edition work? How does it turn spoken word into text and/or program commands? The biggest obstacle to higher levels of accuracy in speech recognition software is that human beings speak in analog "signals" and computers speak an entirely digital language. So for the Dragon transcription software on your PC or Mac to understand what you're saying these analog signals need to be converted into digital data. This conversion is achieved by using a standard microphone, or something more powerful like the Nuance PowerMic II, which takes the analog signals created by your voice and converts them into digital data which a computer can then understand. Now that your computer has digital information to work with it sends that same information to Dragon Medical Practice Edition, which then analyzes the data and converts it into text for your word processor, or alternatively it can interpret spoken commands issued for applications to perform specific functions (like sending an email for example). But how can Dragon Medical transcription software offer such high levels of accuracy? Obviously a big part of this is that when you first set-up Dragon Medical Practice Edition you need to train the software to understand your voice, which can take around 45-minutes. From that point onwards the software can then interpret what you're saying, and the accuracy level rises the more often you use the software itself. The accuracy of all voice recognition software comes down to the fact that any language can be broken down into a given number of basic sounds, which are called phonemes. In the case of the English language it's made up of roughly 50 of these phonemes. The inner workings of any Dragon medical voice recognition software then quickly calculates what it thinks you were trying to say, compares this against its own internal dictionary of several hundred thousand words, and displays the results on screen for you. The more often you use this software the more accurate it becomes – it’s really that simple. Dragon Medical Practice Edition is the leading voice recognition software for medical transcription on the market today, and it's also available in a Spanish language edition.
Related Articles -
Dragon, Medical, Practice, Edition,
| 
