"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 fuck yourself you piece of shit!!!!" 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.
QuoteI'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.