By PYMNTSPosted on October 15, 2018
Amazon is trying to make Alexa, its voice-activated personal assistant, act as a doctor or nurse, detecting illness by a change in the user’s voice.
The Next Web reported the company recently filed a patent in which Alexa can determine different things about a person based on their voice. For instance, it can determine if someone is happy, sad, mad, bored or afraid by the sound of their voices, and can determine based on the user’s breath or voice if they are crying or tired. The patent also covers being able to identify the user based on accent, age and gender, as well as background noise.
The report noted that with the data, Amazon plans to send ads to users based on how they are feeling. For instance, if Alexa was able to determine that a user has a cold, it could present ads for cough medicine. The Next Web noted that Amazon has been keenly interested in understanding the behavior of its users and enabling Alexa to react accordingly. The report pointed to Whisper, a recently launched feature in which Alexa will whisper back an answer when the question is posed in a lower voice. And in June, the company inked a deal with the National Health Service in the U.K. to diagnose symptoms for medical conditions via Alexa.
With the goal of Alexa being embedded everywhere, Alexa announced in September the launch of a new API dubbed Alexa Gadget Toolkit, which is aimed at creating more functions for Alexa gadgets. According to Amazon, Alexa gadgets are accessories that pair with compatible Echo devices via Bluetooth, extending Alexa’s abilities with motors, lights and sound chips, among other things. The toolkit offers developers self-service APIs, including gadget interfaces that provide the metadata of Alexa capabilities on compatible Echo devices. The toolkits also include technical documentation and sample code that enables direct pairing and connectivity, communication and over-the-air updates between the gadget and the paired Echo device. Amazon noted that building on Alexa Gadget doesn’t require advanced processors, microphones or audio processing.