Amazon’s new Alexa will ask you follow-up questions



[ad_1]

Illustration from the article titled Amazons Alexa Can Now Ask You Follow-Up Questions

Photo: GRANT HINDSLEY / Contributor (Getty Images)

Responding to consumer requests for “an Alexa, but make her talk even more,” Amazon announced Wednesday that its latest digital assistant model will be able to deduce a user’s “latent goals” and use them to ask follow-up questions to users.

If you ask your new Alexa how long it takes to cook an egg too easy, Alexa will tell you it takes roughly 1 1/2 minutes – then ask if you want to set a timer.

According to Amazon, the update involved algorithmic adjustments and a deep learning model which will allow the device to evolve according to its relationship with the user. If you ask Alexa how to cook eggs every morning and still choose to set the timer, Alexa discovery model will use active learning to improve your predictions and determine more precisely whether or not you want to know when these 1 1/2 minutes have passed.

“Amazon’s goal for Alexa is for customers to find interacting with her as natural as interacting with another human being,” Amazon wrote in a blog post. “While [apps] may have different results, our first measurements show that the latent objective [inference] increased customer engagement with some developers’ apps. “

Amazon seems to recognize the fact that discovering latent lenses has the potential to be very, very boring. In the first prototypes, when requested users “Chicken recipes”, for example, Alexa would have Continue by asking, “Do you want me to play chicken sounds?”

In order to mitigate the potential for unwanted suggestions, Amazon has implemented a trigger model based on deep learning which takes into account “several aspects of the dialogue context, such as the text of the client’s current session with Alexa and whether the client has engaged with Alexa’s multi-skill suggestions in the past. “

Latent goal inference follows Amazon launch Alexa Conversationsa series of deep neural networks aiming to manufacturing it is easier for developers to integrate a natural conversational experience in custom apps – and is currently available in English for the United States.

[ad_2]

Source link