Method and system for converting Chinese characters into general braille alphabets

Abstract

The invention provides a method and system for converting Chinese characters into general Braille, and the method comprises the steps: constructing a deep learning model which is provided with two sub-task branches sharing feature codes, one sub-task branch is used for carrying out the word segmentation of a Chinese text according to a Braille rule, and the other sub-task branch is used for carrying out the feature coding; the other subtask branch is used for labeling pinyin and corresponding tones for each Chinese character; alternately training two subtask branches of the deep learning model by using a Chinese character corpus marked with braille segmented words and a contrast corpus of Chinese character pinyin until loss functions of the two subtask branches are converged or reach a preset number of iterations, and storing the current deep learning model as an identification model; and inputting Chinese character data to be subjected to general Braille conversion into the recognition model to obtain word segmentation information and pinyin information with tones of the Chinese character data, and converting the Chinese character data into a corresponding general Braille text based on the word segmentation information and the pinyin information with tones. According to the invention, real-time accurate conversion from Chinese characters to braille alphabets can be realized.

Description

technical field [0001] The invention relates to the technical field of Braille processing, in particular to the field of automatic conversion from Chinese characters to common Braille. Background technique [0002] Braille is an important way for blind people to read and obtain information. It is a tactile symbology, printed on paper or displayed on a dot display, that is read by touch. The basic unit of Braille is called "square". One side contains 6 dots. By setting whether each dot is dotted or not, a total of 64 combinations can be formed (including an empty square with no dots in 6 dots). These combinations constitute the most Basic braille symbols. [0003] In order to generate Braille content, textual content used by ordinary people needs to be converted into Braille. For alphabetic scripts, there is a direct mapping from letters to Braille symbols, and the conversion is relatively simple. Currently, computer systems are available for automatic conversion of Engli...

AI Technical Summary

Problems solved by technology

[0011] The traditional method generally performs the above three steps in sequence, but there are two problems in this method: one is that the time required for each step needs to be superimposed, resulting in more time required overall; the other is that the latter step depends on the previous step , if the result of the previous step is wrong (such as word segmentation error), it will affect the subsequent steps (such as pinyin marking and standardization), resulting in error superposition

However, there is still a huge difficulty in this idea for general Braille: if you want to use a model to realize word segmentation and pinyin marking at the same time, you need a Chinese character-common Braille comparison training corpus, that is, a large amount of corpus is required, and these corpus Requires both Braille word segmentation and Pinyin marking information

At present, because the general Braille is in the promotion period, there are few corpus available, and it is difficult to meet this requirement

Images