Speech Dictation System Based on Character Recognition  

作　　者：Wenjun Lu Yanqing Wang Longfei Huang 

机构地区：[1]College of Information Engineering,Nanjing Xiaozhuang University,Nanjing 211171,China

出　　处：《国际计算机前沿大会会议论文集》2021年第1期380-392,共13页International Conference of Pioneering Computer Scientists, Engineers and Educators（ICPCSEE）

基　　金：This article is supported by the 2020 Innovation and Entrepreneurship Training Program for College Students in Jiangsu Province(Project name:Mom doesn’t have to worry about my dictation any more-dictation software based on character recognition,No.202011460104T);This article is supported by the National Natural Science Foundation of China Youth Science Foundation project(Project name:Research on Deep Discriminant Spares Representation Learning Method for Feature Extraction,No.61806098);This article is supported by Scientific Research Project of Nanjing Xiaozhuang University(Project name:Multi-robot collaborative system,No.2017NXY16).

摘　　要：To solve students’ dictation problems, a speech dictation system basedon character recognition is proposed in this paper. The system applied offlinehandwritten Chinese character recognition technology, denoised the imagethrough Gaussian filter, segmented the text through projection method, and convertedthe image to text through OCR technology. The straight line mark in thepicture was detected by Hough transform technology, and then SKB-FSS algorithmand WST algorithm were used for speech synthesis. Experiments show thatthe system can effectively assist students in dictation.

关 键 词：Character recognition Speech synthesis Hough transform Feature extraction Image preprocessing 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]