Technical articles

Why implement a chatbot with Amazon Lex?

What is a chatbot?

Technological progress has changed the way people communicate. According to the report of Gartner “Chatbots Will Appeal to Modern Workers 2019”, By 2022, 70% of workers will interact with conversation platforms on a daily basis. This growth is due to chatbots meeting the demand for instant digital connections that keep users updated at all times. As a result, it is currently a trend that organizations are investing in the development and implementation of a chatbot. A chatbot is a software that simulates and processes a spoken or written conversation on different chat platforms such as WhatsApp, Messenger, Slack, etc. Its objective is not to replace a human, but to help in the most natural way to solve repetitive tasks such as answering customer inquiries in a support service.

 

The main benefits of having a chatbot are the following:

  • It automates business processes by enabling customers to be served immediately and significantly improves the user experience.
  • Reduce costs by not having 24-hour agents answering questions. 
  • It maintains an active costumer engagement by making it possible to contact costumers whenever we want, because 75% of smartphones have a messaging application.
¿What is a chatbot?

We think that a good option is to implement a chatbot with Amazon Lex, since Amazon Lex is an Amazon Web Services (AWS) service. AWS is the most widely adopted and comprehensive cloud platform in the world, offering more than 175 comprehensive data center services globally. (Amazon, 2020).

Amazon Lex is a service for creating text and voice conversational interfaces in any application. Amazon Lex offers the advanced deep learning capabilities of automatic speech recognition to convert speech to text and natural language understanding technology to recognize text intent, allowing you to create applications with highly interactive user experiences and realistic conversations.

Amazon Lex puts Amazon Alexa deep learning technologies in the hands of any developer, allowing you to quickly and easily create sophisticated natural language chatbots (“chatbots”) . As a fully managed service, Amazon Lex scales automatically, so you don’t have to worry about managing infrastructure. “

Amazon, 2020

Amazon Lex allows you to implement chatbots that can be integrated with channels such as Messenger, Twilio, and Slack. In addition, Lambda functions can be invoked giving the possibility of integrating with more AWS services as well as making requests to external APIs. For the use of Amazon Lex, AWS charges a competitive rate that is calculated based on the number of text or voice requests processed by the bot that are added at the end of the month to determine the monthly charges. Current prices are: $ 0.004 per voice request and $ 0.00075 per text request. So for a bot that processes 1,000 voice requests and 1,000 text requests in a given month it would cost $ 4.75 (Amazon, 2020).

Case

At bit2bit Americas we implement a chatbot capable of searching for content in Confluence from Slack with the architecture shown in the image.

Using Amazon Lex allowed the chatbot to understand and respond to messages in Spanish, it also facilitated the integration between Slack and Amazon Lex. It also allowed you to interact with the Confluence API from Lambda functions.

Conclusion

In conclusion, currently it is key to have a chatbot that interacts with customers due to the benefits it offers to organizations and a good option is to implement it with Amazon Lex due to the integrations it offers and its low cost.

¿Por qué implementar un chatbot con Amazon Lex?
Comparte esto:

Noticias relacionadas

Technical articles

Technological progress has changed the way people communicate. According to the Gartner report “Chatbots W ...

Technical articles

With the demands of today's infrastructure, it is crucial that organizations implement automation of ...

Technical articles

Jira is a powerful tool - it's great for planning, tracking, and managing projects. But sometimes it doesn't allow you to do exactly what ...