The lowly Arduino, an 8-bit AVR microcontroller with a pitiful amount of RAM, awfully little Flash storage space, as well as efficiently no peripherals to speak of, has much better speech recognition abilities than your Android or iDevice.  Eighty percent accuracy, compared to Siri’s sixty.Here’s the video to show it.

This uSpeech library produced by [Arjo Chakravarty] utilizes a Goertzel algorithm to turn input from a microphone linked to one of the Arduino’s analog pins into phonemes. From there, it’s fairly simple to turn these caught phonemes into function requires illumination a LED, turning a servo, or even replicating the Siri, the contemporary version of the Microsoft paperclip.

There is one caveat for the uSpeech library: it will only respond to predefined phrases as well as not typical speech. Still, that’s an incredibly outstanding achievement for a easy microcontroller.

This isn’t the very first time we’ve seen [Arjo]’s uSpeech library, however it is the very first time we’ve seen it in action. When this was published months as well as months ago, [Arjo] was behind the fantastic Firewall of China as well as couldn’t publish a appropriate demo. because this the uSpeech library is a spectacular accomplishment we requested a few videos showing off a few applications. nobody made the effort, so [Arjo] chose to make utilize of his new VPN as well as show off his work to the world.

Video below.