Chinese grammatical error recognition method, device, computer equipment and storage medium
A grammatical error and recognition method technology, which is applied in computing, natural language data processing, unstructured text data retrieval, etc., can solve the problems of low recognition rate of Chinese grammatical errors, Chinese grammatical errors, etc., and achieve the effect of improving accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0046] An embodiment of the present invention provides a method for identifying grammatical errors in Chinese, which identifies grammatical errors in Chinese speech. Through this method, the accuracy of identifying Chinese grammatical errors can be improved. Specifically, figure 1 The flow chart of the steps of the recognition method for Chinese grammatical errors provided by Embodiment 1 of the present invention, such as figure 1 As shown, the method for identifying Chinese grammatical errors provided by this embodiment includes the following steps S101 to S106.
[0047] Step S101: Obtain speech words to be recognized.
[0048] Wherein, the method for identifying Chinese grammatical errors can be performed on a single speech, or on a text including multiple speeches such as an article or a paragraph of text. When identifying a single speech, the single speech is the speech to be identified. When recognizing the text, the words can be separated first, and then each separated...
Embodiment 2
[0085] Embodiment 2 of the present invention provides a preferred method for identifying Chinese grammatical errors. Part of the features and corresponding technical effects are the same as those in Embodiment 1 above. Reference may be made to Embodiment 1 above, and details will not be repeated here. In this embodiment, the M model groups corresponding to the one-to-one correspondence between the bidirectional cyclic neural network and the LR classifier are preset, that is, a plurality of model groups are trained. Each model group can obtain a recognition index of a speech to be recognized, that is, each speech to be recognized can obtain M recognition indexes, where M is a natural number greater than 1. specifically, figure 2 The flow chart of the steps of the recognition method for Chinese grammatical errors provided by Embodiment 2 of the present invention, such as figure 2 As shown, the method for identifying Chinese grammatical errors provided by this embodiment inclu...
Embodiment 3
[0099] Embodiment 3 of the present invention provides a preferred method for identifying Chinese grammatical errors. Part of the features and corresponding technical effects are the same as those in Embodiment 1 above. Reference may be made to Embodiment 1 above, and details will not be repeated here. This embodiment is aimed at the recognition of the text to be recognized including multiple speeches. In this embodiment, M model groups corresponding to one-to-one correspondence between the bidirectional cyclic neural network and the LR classifier are preset, that is, multiple models are trained. The structure of each model group is the same, the initial training parameters are randomly generated, and the corresponding wrong speech group of the text to be recognized can be obtained through each model group, that is, the text to be recognized corresponds to M wrong speech groups, where , M is a natural number greater than 1. specifically, image 3 The flow chart of the steps of...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


