The invention discloses a method for correcting the error characters in theresult of speechrecognition and a speech recognition system. The method comprises thefollowing steps:marking the error characters in the result of speech recognition; inputtingthe correct characterscorresponding to the marked error characters by the input based on characters-shape; recognizingthe stroke-based input; displaying the candidate characters; selecting thedesired characters fromthe candidate characters by the user; and replacing the error characters withthe selectedcharacters, said method characterized by further comprising steps: filteringthe candidatecharacters by means of the acoustic information of the error characters.
展开▼
