Researchers build an AI that learns equally well from visual, written or spoken materials

Share

Researchers build an AI that learns equally well from visual, written or spoken materials

What Happened?

Advances in the AI realm are constantly coming out, but they tend to be limited to a single domain: For instance, a cool new method for producing synthetic speech isn’t also a way to recognize expressions on human faces. Meta (AKA Facebook) researchers are working on something a little more versatile: an AI that can learn capably on its own whether it does so in spoken, written, or visual materials. Advances in the AI realm are constantly coming out, but they tend to be limited to a single domain: For instance, a cool new method for producing synthetic speech isn’t also a way to recognize expressions on human faces. Meta (AKA Facebook) researchers are working on something a little more versatile: an AI that can learn capably on its own whether it does so in spoken, written, or visual materials.

Read the full article: https://tcrn.ch/3Il9hmc

Why is this important?

I'm excited about the potential for artificial intelligence (AI) programs that can learn equally well from visual, written, or spoken materials. This would be a huge step forward in terms of AI development and AI learning and could lead to more accurate translations, improved search results, and better comprehension of instructions given by humans. I'm looking forward to seeing this technology come out of the lab and be used in everyday applications that make technology more accessible for human beings and therefore make our society a little bit more humane!