Voice Driven Bot For Cross Domain Database Querying

Affra H.¹ , Nageshwari S.² , Poorvika A.N.³

Section:Research Paper, Product Type: Journal-Paper
Vol.7 , Issue.6 , pp.28-35, Jun-2021

Online published on Jun 30, 2021

Copyright © Affra H., Nageshwari S., Poorvika A.N. . This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
 

View this paper at   Google Scholar | DPI Digital Library

XML View     PDF Download

How to Cite this Paper

Abstract :
Many applications in health, medicine, finance store their information in relational database. Users cannot precisely work with structured query languages like SQL unless they have strong prior knowledge in this field. Also SQL is difficult to expertise for non­technical users. Hence a long­standing goal is to allow the users to interact with database with natural language. The reason for Voice­Driven bot is it can also be an assistive technology for visually impaired people. The system involves 3 main phases : Speech to text, Text to SQL queries and Text to speech. In the First phase,the input is received as voice signals which will be used to predict the text from the audio file. Then in the second phase, SQL queries are generated from the text using encoder­ decoder mechanism and pick the one which is valid and less complex to fetch the results. The model is trained using the Spyder dataset which makes the model aware of relations between the tables. After that, the results are converted to complete sentence and delivered back as a voice reply to the user. The natural language query from the user is converted to text using Speech recognition. Deep Learning is used to train the neural networks on large scale data of questions and answers.Bridge model finds the table names, column names and the conditional operators. SQLite Database is used to fetch the results based on the generated query

Key-Words / Index Term :
Speech Recognition, SQL, BRIDGE model, NLP

References :
[1] Ahmed Elgohary, Saghar Hosseini, Ahmed Hassan Awadallah, ”Speak to your Parser: Interactive Text­to­SQL with Natural Language Feedback”, arXiv:2005.02539v2 ,2020.
[2] Bailin Wang, Richard Shin,Xiaodong Liu, Oleksandr Polozov, Matthew Richardson, “RAT­SQL: Relation­Aware Schema Encoding and Linking for Text­to­SQL Parsers”, In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, 2020.
[3] Dario Amodei, Rishita Anubhai, Eric Battenberg, Carl Case, Jared Casper, Bryan Catanzaro, Jingdong Chen, Mike Chrzanowski, Adam Coates, Greg Diamos, Erich Elsen, Jesse Engel, Linxi Fan, Christopher Fougner, Tony Han, Awni Hannun, Billy Jun, Patrick LeGresley, Libby Lin, Sharan Narang, Andrew Ng, Sherjil Ozair, Ryan Prenger, Jonathan Raiman, Sanjeev Satheesh, David Seetapun, Shubho Sengupta, Yi Wang, Zhiqian Wang, Chong Wang, Bo Xiao, Dani Yogatama, Jun Zhan, Zhenyao Zhu , “Deep Speech 2: End­to­End Speech Recognition in English and Mandarin”, 2015.
[4] DongHyun Choi, Myeong Cheol Shin, EungGyun Kim, Dong Ryeol Shin, “RYANSQL: Recursively Applying Sketch­based Slot Fillings for Complex Text­to­SQL in Cross­Domain Databases”, Association for Computational Linguistics, 2020.
[5] Jiaqi Guo, Zecheng Zhan, Yan Gao, Yan Xiao, Jian­Guang Lou, Ting Liu, Dongmei Zhang “Towards Complex Text­to­SQL in Cross­Domain Database with Intermediate Representation”, 2019.
[6] Jichuan Zeng, Xi Victoria Lin, Caiming Xiong, Richard Socher, Michael R. Lyu, Irwin King, Steven C.H. Hoi, “Photon: A Robust Cross­Domain Text­to­SQL System”, 2020.
[7] Prerana Das, Kakali Acharjee, Pranab Das and Vijay Prasad, “Voice Recognition System : SPEECH­TO­TEXT”, International Journal of Applied and Fundamental Sciences,pages 191­195, 2016.
[8] Puneet Kaur, Bhupender Singh,Neha Kapur, “Speech Recognition with Hidden Markov Model”, 2014.
[9] Tao Yu, Michihiro Yasunaga, Kai Yang, Rui Zhang, Dongxu Wang, Zifan Li, and Dragomir Radev, “Syntaxsqlnet: Syntax tree networks for complex and cross­domain text­to­sql task”, In Proceedings of the Conference on Empirical Methods in Natural Language Processing pages 1653–1663,2018.
[10] Tong Guo ,Huilin Gao “Content Enhanced BERT­based Text­to­SQL Generation ”,arXiv:1910.07179v5 , 2020.
[11] Xi Victoria, Lin Richard, Socher Caiming Xiong,”BRIDGE : Bridging Textual and Tabular Data for Cross­Domain Text­to­SQL Semantic Parsing”, 2020.
[12] Xiaojun Xu, Chang Liu, Dawn Song “ SQLNet: Generating Structured Queries From Natural Language Without Reinforcement Learning”, 2017.