AWS LEX NLP - Improve Voice Recognization Accuracy

Register
Submit a solution
The challenge is finished.

Challenge Overview

We previous build several AWS Lex Intents for our client, but the voice recognization accuracy is very poor. 
The purpose of this challenge is to improve the voice recognization accuracy to be higher than 70%. 
There are several intents in the provided codebase, but you can work on any one (only one is enough) to improve the accuracy. 

The codebase git repo is posted in the contest forum. The winner is required to push your submission to the git repo. 

Note to reviewer: You have to a be native English speaker to test the submission properly. Please consider this when apply the review position. 
The submission can only be accepted if its voice recognization accuracy is higher than 70%. 

Final Submission Guidelines

Submission Deliverable
- Updated Source Code
- Easy to follow Deployment Guide and Verification Guide

Note that you can choose to use another language (other than Node.js) if necessary. 

ELIGIBLE EVENTS:

2018 Topcoder(R) Open

Review style

Final Review

Community Review Board

Approval

User Sign-Off

ID: 30064531