Keyword correction method and device, computer equipment, and storage medium

Abstract

The invention relates to a keyword correction method and device, computer equipment, and a storage medium. The method comprises the following steps: screening out a plurality of reference texts related to keywords of a to-be-processed text from a memory, and determining to-be-corrected keywords in the to-be-processed text according to the plurality of reference texts; determining candidate reference texts from the plurality of reference texts according to similarity between the to-be-corrected keywords and the plurality of reference texts; determining a target reference text according to phoneme similarity between the to-be-corrected keywords and the candidate reference texts, and finally correcting the to-be-corrected keywords based on the target reference text. According to the keyword correction method provided by the invention, the target reference text is determined according to different phoneme similarity of the candidate reference texts and the to-be-corrected keywords from phoneme strength; it is ensured that the editing distance between the to-be-corrected keywords and the candidate reference texts with high similarity is smaller than the editing distance between the to-be-corrected keywords and the candidate reference texts with low similarity, so that the correctness of correcting the to-be-corrected keywords can be improved.

Description

technical field [0001] The present application relates to the technical field of speech recognition, in particular to a keyword correction method, device, computer equipment and storage medium. Background technique [0002] Speech recognition technology converts audio into text through intelligent recognition equipment. With the remarkable progress of speech recognition technology, speech recognition technology has gradually entered various fields such as automotive electronics, medical care, finance, and consumer electronics. However, due to internal reasons such as each person's birthplace and pronunciation habits, as well as external reasons such as signal interference and poor network, the actual use accuracy of speech recognition is low, which greatly affects the business that needs to use speech recognition for follow-up operations and work. [0003] The main direction of the existing technology is the tuning and improvement of speech recognition. The recognition algo...

AI Technical Summary

Problems solved by technology

However, in many cases, it is not that the recognition ability is insufficient. Under the technical background that the existing standard Mandarin recognition rate can almost be accurately recognized, the reason for the recognition deviation is due to the difference in human pronunciation and the interference caused by the environment. These problems can only be improved Recognition ability is difficult to overcome or the room for improvement is extremely limited

Therefore, it is not effective in solving the problem of low speech recognition accuracy.

Images