As it stands, devices like Amazon's Echo and Apple's HomePod react to a command given by the user, but can't yet tell the difference between two people talking in the background, or radio static.
Cambridge startup company Audio Analytic is looking to change this and allow machines to really recognise the way we talk for the first time.
As reported by Forbes, the firm announced this week it is adding to its sound profiles with one for human speech, which could mean devices would eventually be able to understand whether to interrupt or not, rather than simply reacting to a word being said.
CEO Chris Mitchell said: "If you look at the speech world, most of speech science is structured around understanding the order in which we're going to say words.
"We control how sound comes out of our mouths. How we do that in speech is called phonemes.
"What Audio Analytic has done is discerned, then written software at the AI level to model ideophones. These are the fundamental building blocks that make up sounds."