Building a Multi-Language Chatbot with Automated Translations
Creating a multi-language chatbot MVP is an easy hit with AWS Lex, Translate, Lambda and API Gateway. Find out how!
Virtual Assistants are here to stay. Regardless of the limitations and complaints they faced in the early years of adoption, many large businesses and banks invested on them. Business Insider forecasted that up to 80% of big firms will implement a Virtual Assistant by next year. Major consulting firms like Deloitte, EY and Accenture are betting on it. Samsung recently joined the smart speaker race with its upcoming Galaxy Home.
I have implemented several chatbots, Skills and Actions and learned that said limitations are most often due to the lack of proper conversation flow design, live testing and iterations. Virtual Assistants are usually fitted with a limited set of dialogue options because of time and budget constraints. The technology is also still facing challenges and it will take time before a conversation with an AI feels as smooth and real as a human interaction. If you are willing to put in the time and effort to cover as many conversation paths and utterances as possibile, though, you can create great experiences with what we have now.
