Smart speakers are always listening out for its activation phrase, whether it be “Hey Siri”, “OK Google”, or “Alexa”. This has created problems where watching a video with those phrases might accidentally trigger your smart speaker, but thanks to the work of researchers, they’re developing AI that could make smart speakers, well, smarter.
This is done by using AI to help determine the direction of the voice that the device is hearing. For example, if you’re far away, the AI might assume that you don’t want to activate the speaker and will ignore those phrases. However, when you’re up close and personal, it assumes that you do want to activate the speaker’s digital assistant.
According to the researchers, “Our approach innately enables voice commands with addressability (i.e., devices know if a command was directed at them) in a natural and rapid manner. We quantify the accuracy of our implementation across users, rooms, spoken phrases, and other key factors that affect performance and usability. Taken together, we believe our DoV approach demonstrates feasibility and the promise of making distributed voice interactions much more intuitive and fluid.”
That being said, we’re not sure if there are plans to actually implement this technology into existing smart speakers, but it could solve a lot of problems and make smart speakers more intuitive in the future.