Correction system for punctuation marks
A punctuation mark and correction system technology, applied in the field of punctuation mark correction system, can solve the problems of inability to meet the changing punctuation mark prediction requirements, low recall rate of neural network model, weak generalization ability, etc., and achieve good generalization ability. and generality, reduce the amount of training data, and improve the effect of accuracy
Pending Publication Date: 2021-04-16
SHANGHAI XIAOI ROBOT TECH CO LTD
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, the neural network model obtained by the existing training method has low recall rate, weak generalization ability and poor versatility, which cannot meet the changing needs of punctuation prediction
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0030] As mentioned above, although the accuracy of the punctuation mark addition method has been greatly improved under the use of neural networks, however, the punctuation mark addition method needs to prepare a large amount of training data in advance for neural network model training, and these training data are often obtained through The non-punctuation corpus generated by Automatic Speech Recognition (ASR) requires time-consuming and laborious manual labeling before training, and then uses the trained neural network model to punctuate the non-punctuation text obtained from speech recognition. Forecasting, the obtained punctuation prediction results tend to have a low recall rate. In addition, the training data of the current neural network model often only considers the information of the word before the punctuation mark, resulting in a very uneven label distribution of the training data. The neural network model trained by this method has poor generalization ability and ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
Disclosed is a correction system for punctuation marks. The system comprises a punctuation mark processing unit suitable for performing punctuation mark removal on a text to be corrected to obtain a text to which punctuation marks are to be added; a punctuation mark labeling unit suitable for inputting the text to which the punctuation marks are to be added into a punctuation mark labeling model which completes transfer learning training, and predicting the to-be-added positions of the punctuation marks in the text to which the punctuation marks are to be added and the types of the corresponding punctuation marks by adopting the punctuation mark labeling model, labeling by adopting a label combination corresponding to the punctuation mark type at front and back word segmentation units of the punctuation mark to-be-added positions, and outputting a corresponding punctuation mark labeling result; and a punctuation mark adding unit suitable for adding a corresponding punctuation mark to the text to which the punctuation mark is to be added according to the punctuation mark labeling result to obtain a corresponding corrected text. According to the scheme, the punctuation mark prediction accuracy can be improved, and the punctuation mark correction requirement is met.
Description
technical field [0001] Embodiments of the present invention relate to the technical field of computer natural language processing, and in particular to a correction system for punctuation marks. Background technique [0002] Existing punctuation recovery schemes usually use sequence labeling, which is mainly applied to recovering punctuation marks in text obtained from speech recognition. Generally, only simple punctuation marks can be added, such as commas and periods. The punctuation added in this way, on the one hand, has low accuracy and poor generalization ability; on the other hand, the marked punctuation marks are poor in richness, resulting in a poor reading experience. [0003] With the continuous development of deep learning technology, the trained and learned neural network model can be used to predict the punctuation marks of the text obtained from speech recognition and improve the accuracy. However, the neural network model obtained by existing training method...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/33G06F40/117G06F40/205G06N3/04G06N3/08
Inventor 沈大框陈培华陈成才
Owner SHANGHAI XIAOI ROBOT TECH CO LTD