Dan Borufka

Natural Language Generation for AI Assistants: Today and in the Future

With current technologies, machines are able to generate natural sentences from data. While engines like SimpleNLG allow us to realise properly inflected forms when changing grammatical parameters, others are able to summarize the important parts of long text bodies. Probabilistic sequence-to-sequence models can construct responses word by word, yet most assistants today are bound to a painstaking traditional process of manually crafted text templates in order to maintain the persona of the company they represent. Let us imagine for a moment what personalities could form once we break out of these confinements and push the boundaries of these technologies...

Having spent 16 years in the IT and marketing industry, Dan's work consisted of cross-functional team management, software development, user-centered design, and language-driven products. He spent 4 years studying and working in California, has worked freelancing, for small businesses and global corporations, lead and trained teams, and joined a startup on Bali. Currently, Dan is making machines talk for Samsung through building the natural response system of their flagship voice assistant Bixby.

Buttontwitter Buttonlinkedin
This website uses cookies to ensure you get the best experience. Learn more