Image for post
Image for post

This is what my Echo Show apparently heard yesterday after a false trigger. What’s more, this was the response to my nearly three year old asking “What’s that?” Apparently, Alexa has a lot of things on her mind lately.

This prompted a very clear and loud “ALEXA, STOP!” response from me. What’s great is that the Echo Show was also displaying the text of the response (thankfully, no visuals).

This brings up some interesting content issues. The first is that the Echo, being a family-oriented device, has to be an arbitrator of what is appropriate content and the range of this can vary widely between different families. The second is adding a higher confidence requirement when to certain content is being asked. “What’s that” vs “What’s sex” is a of a high WER but if internally, Amazon must have its own confidence scoring for the STT result. Maybe this was in the 70% range (just a guess)? Maybe for request that involve “sex” or other potentially adult requests, the confidence level needs to be higher?

Also, what might be useful down the line is to add a child voice detector. Perhaps Alexa could zero requests that are originating from child-sounding commands?

Independent daily thoughts on all things future, voice technologies and AI. More at http://linkedin.com/in/grebler

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store