Imitate nature? (Score 1)

When you want to talk to another person you tend to turn towards them and speak, I expect people would behave the same way with a computer. Why don't hardware developers design computers with stereo mics that's a little apart like human ears, and use a little signal processing circuitry so the computer can work out if it's being spoken to or not? I don't know how feasible that is, but it sounds like a simple thing to do compared to getting software to learn a person's voice characteristics.

Also why do these app developers think that voice command *necessarily* need to consist of commands in plain English (or any other human language)? I really want some sort of plugin for my media player that'll allow simple commands to be entered by whistling, for example say I want it to play, instead of saying in a flat monotone voice 'computer, play' or something silly, I clap twice (to get the computer's attention), and whistle two notes a major third apart, another command can then be a fourth, a fifth etc (so the user doesn't need to have perfect pitch). That would be far more reliable as command input, the user won't sound like a wally, and easier(?) to implement in software.

Pls pls any talented developers here take this idea up!