Why is it difficult?

Conversational AI is a technology that allows humans to communicate with a computer and establish meaningful conversations. A crucial requirement for such an AI-powered device is the ability to recognize and comprehend human language. Advances in machine learning and natural language processing have already delivered voice recognition and text translation for many languages. The challenge arises when individuals converse naturally with mixed language input from 2-3 languages. In addition, emotions, accents, and contextual intonation make it even harder for conversational AI to interpret the intended user meaning and respond appropriately. Training machine learning models to meet these requirements would require a large sample of data for each combination of language and context, which is hard to obtain.

What is the impact?

Solving these challenges would allow AI-powered personal assistance devices such as Alexa or smart-watches to perform seamlessly for billions of people worldwide. More people would benefit from access to the internet in their language, being able to perform tasks ranging from research to online healthcare to customer support and so much more. Conversational AI holds the promise of democratizing information access for all, particularly overcoming issues of varying literacy levels and offering a more natural interface to the new internet users.