Intelligent super market assistant with speech and voice
Supermarkets are being popular among all types of people and larger in size in modern society. People prefer supermarkets over groceries because they can choose what they want, can check and compare prices before buying and a wide variety of products are available at supermarkets. Despite this popularity, assistants at many supermarkets are inefficient and don't meet the customer requirements. They don't very much aware of the products available at the supermarket, some of them don't express a pleasant look towards the customers and reluctant to help the customers. 

Lei is a successful technological solution to above problems of human super market assistants. Lei is capable of providing intelligent answers to customer queries about the product of supermarket in form of voice or text. Customers can also use voice commands or text to communicate with Lei.
What is Lei?

Lei is placed at several locations of a supermarket with a display to visualize the results, a microphone to receive voice queries from the customer, speaker for speech and keyboard to input queries by text. Customers can check the availability of products, available product types and brands, the location where the product has been stored in the supermarket, price comparisons of the same product in different brands, available offers and much more. The most important thing about the Lei is customers don't have to use specific questions types or command words to communicate with Lei. We are doing natural language processing to understand the user input and therefore users can just ask questions as he is asking from a human assistant. 

Lei is efficient, accurate and smart than human assistants. By placing enough Lei access points inside a supermarket, we can completely remove all human assistants except cashiers from the supermarkets. Lei is actually cheaper than paying salaries to human assistants. It only costs the initial installation cost which is again not much higher. 

How does Lei work?

Any supermarket contains a sophisticated database about their products stock. The Lei is connected to that database and we implement a database mapper to extract the required information from that database in a format which is needed for Lei. Then those extracted data are processed and an index is built in order to answer for natural language queries. We are using Microsoft voice recognition and speech to text APIs of Microsoft cognitive services to recognize voice inputs and reply to the customer with voice. Stanford CoreNLP is used for natural language processing. 

Our product Lei won the 1st Runners up of Google IO extended Hackathon 2016. Then we were invited by Dialog Ideamart to continue this project as a startup with investments. 

Project Repository :-  https://github.com/DVios/lei
  1. Managing Director
  2. Managing Director
  3. Managing Director
  4. Managing Director
  5. Managing Director