| CPC G10L 17/14 (2013.01) [G06F 3/04842 (2013.01); G10L 17/02 (2013.01); G10L 17/22 (2013.01); G10L 25/57 (2013.01)] | 15 Claims |
|
1. A method of editing a speech recognition result, the method being performed by a computing device including at least one processor,
the method comprising:
displaying a word list satisfying a predetermined condition based on text information generated by speech recognition, wherein the word list is displayed adjacent to the text information;
performing at least one of an action of determining a target word within the word list or an action of receiving input of a target word through a user interface;
displaying a region corresponding to the target word in the text information;
determining a dynamic threshold for the target word based on a length of the target word;
calculating an edit distance between the target word and an additional search target word;
determining that the additional search target word is similar to the target word based on the edit distance being equal to or less than a dynamic threshold;
searching for both the target word and the additional search target word in the text information; and
after searching for both the target word and the additional search target word, changing, in bulk, the target word and the additional search target word to an inputted change word,
wherein the predetermined condition includes at least one of predetermined word information for each user account or predetermined threshold information associated with a frequency of occurrence of a word.
|