Font Size: a A A

The Development Of Universal Data Input System Based On Speech Recognition

Posted on:2019-11-22Degree:MasterType:Thesis
Country:ChinaCandidate:X Q ZhaoFull Text:PDF
GTID:2428330575992224Subject:Engineering
Abstract/Summary:PDF Full Text Request
In recent years,with the changes of times and the progress of science and technology,artificial intelligence has developed rapidly,especially the development of speech recognition technology,with its various types of products have become increasingly mature.What it brings to people is more than just abundant resources,but also more convenient ways of operating.However,in the current market,the research related to data collection based on speech recognition is still in its infancy,and there is no mature products that could be put on the market.By using the relevant design ideas of software engineering,this article prepared for the preliminary requirement analysis and system function design through the early investigation of forestry information collection and other related universal data collection work,fully considering the work habits of the data collection staffs and the basic characteristics of data.Then we developed a data input system based on speech recognition,and tested on the unit and function modules.This platform integrates speech recognition input,data storage,and data export functions.Users of the system can customize the collection data table according to their own needs,and then complete the speech recognition input of the data through the customized data table.For the convenience of users,the system also designed the questionnaire including forest and tree information questionnaire,student information questionnaire,and personal physical examination questionnaire.The input mode of the system is divided into two kinds:voice input mode and manual input mode.The voice input mode is the default input mode of the system,and the function of voice recognition is realized by invoking an API interface integrated in the system so that the user can perform voice human-machine interaction;When the input data of speech recognition is inaccurate,the latter mode is enabled,and the system can invoke the system input method of the Android mobile phone to complete the input.In addition,through the integration of the two input modes,the system allows users to complete the input mode switch with one click.In the system information storage section,users can add,delete,modify,and view the collected structured data.Once the input is completed,the imported data can be exported to the computer using the data export,and the data can be generated in the form of an Excel table.After nearly a hundred tests,the system successfully reduced the input times by more than one-fourth compared to the original manual input times,and reduced the frequency of manually-clicked screens by more than 40 times.This proves that the system can effectively improve the input efficiency and reduce the times of user operations.
Keywords/Search Tags:Speech recognition, Speech input, Android, Data collection
PDF Full Text Request
Related items