Dragon Professional v15 offers highly accurate desktop speech recognition out-of-the-box. While Nuance’ s speech recognition engine incorporates deep learning technology it does not have the intelligence or experience of a human being. For Dragon to learn your voice fully it needs to build a statistical model based on voice data as well as manual corrections. When Dragon misrecognizes a word that is spoken the end user must teach the software through the Correction Process the word or phrase it should have transcribed.
The Correction Process is a critical step in improving the voice profile and is a tool to help Dragon match the words you speak with the words it types. This process should be done during or after every dictation to ensure that Dragon learns to recognize the way you speak words.
Deleting the text will not improve your voice profile so make sure you correct misrecognized words before you edit.
VocaLinks Coaching programs teach the Dragon user strategies and best practices for building the accuracy of their Dragon voice profile whether it is for school, work or general hands-free computer use.
Need more details? Please contact us to discuss further.