A speech information processing method includes: determining text information corresponding to collected speech information according to a speech recognition technology, wherein the text information comprises a word; with the word in the text information as a target word, determining one or more fuzzy words corresponding to the target word according to a phoneme sequence corresponding to the target and a preset phonetic dictionary, wherein the phonetic dictionary comprises a plurality of words and phoneme sequences corresponding to the plurality of words; and outputting the target word and the one or more fuzzy words corresponding to the target word.