a) Your phone has a microphone that picks up information spoke more-or-less directly into it. An Echo is specifically equipped with a microphone array designed

a) Your phone has a microphone that picks up information spoke more-or-less directly into it. An Echo is specifically equipped with a microphone array designed to extract audio from anywhere in the room. b) There is a great deal less ambiguity around "pressing buttons" than there is around interpreting speech. While it is unlikely that your phone will incorrectly detect button presses, it's very common for voice-activated devices to a) incorrectly detect a wake word (either negative or positively); and b) misunderstand some particular word used in the command. Your phone is not going to think you pressed the "Clothes" button when you actually pressed "Close". c) The entire functionality of the device is accessible from behind that big ambiguous interface. On a phone, there are many distinct steps and screens to step through when you want to do something (this complex interface, by the way, is a non-trivial part of why butt-dials are quite rare these days). On a "smart speaker", most things are just one misheard statement/command away from occurring.