The Core
Why We Are Here => Water Cooler => Topic started by: Mackin USA on October 15, 2018, 12:24:43 PM
-
“A current physical and/or emotional condition of the user may facilitate the ability to provide highly targeted audio content, such as audio advertisements or promotions, to the user,” Amazon’s patent says.
-
Luckily they already acquired PillPack, so whenever you feel down they'll have the perfect pick-me-up. Soon enough they will advertise free telemedicine to push scripts ;)
-
I think this is already built into some voice recognition software in rudimentary ways.
I figured this out when stuck in a looping phone menu and was trying phrases like "other" and "customer service representative" and pushing zero and looping and looping. Finally I said "Go f### yourself you piece of sh##!!!!" and immediately it came back with "Let me get a customer service representative to assist you."
I've tried it quite a few times since and it has worked in maybe 20% of the cases. Conceivably random, but I'm not so sure. Granted, if it is really happening, I'm sure it's keyword matching, not measuring stress in the voice or "emotion" directly. Still... I kind of believe it's a thing.
-
I'm sure it's keyword matching, not measuring stress in the voice or "emotion" directly.
Probably. A number of machine learning tools offer this, including Alexa IIRC. E.g. IBM Watson's Tone Analyzer: https://www.ibm.com/watson/services/tone-analyzer/
Effectively: Voice > Text > Tone Analyzer > Action. All in real time. Very impressive.