Automatic speech recognition adaptation using user corrections

The device for automatically speech recognition system identifying user to orally accounting the change of the text; and infers device of nanostructures for source automatic the user to his/her mind, or the change for is a identifying result device. Detecting and identifying to provide, the type of which the system using user for methods and apparatus for modifying, itself to reduce the opportunity of a recognition which error program. Therefore, the system and method for provided at least one or capillary needle with the extra user interactive salient significant voice identifying learning.