Speech recognition error correction method, system, device and storage medium

The invention discloses a voice recognition error correction method, a system, a device and a storage medium, and relates to the artificial intelligence technology, and the method comprises the steps: carrying out the voice recognition of a to-be-detected voice, and obtaining a to-be-detected text and a corresponding to-be-detected pronunciation sequence; constructing an FST to be detected according to the pronunciation sequence to be detected; according to the to-be-detected FST and the keyword FST, determining a plurality of to-be-corrected words in the to-be-detected text, and determining to-be-corrected statements containing the to-be-corrected words; if the to-be-corrected words exist in the Chinese character confusion set, determining a replacement word corresponding to each to-be-corrected word, and replacing the to-be-corrected words in the to-be-corrected statement by the replacement words to generate a replacement statement; when the first logic score of the to-be-corrected statement is smaller than