This morning when I woke up to the sound of the alarm I set on my Amazon Echo last night, I thought to myself "I wish more of my gadgets worked by me yelling things at them." Luckily, a few minutes later I was reading a PR email from SeeedStudio about its new product ReSpeaker which just launched on Kickstarter and promises to turn everything in your house into an Amazon Echo.
ReSpeaker is a hardware interface that lets you add voice control to anything, as long as you’re willing to put in the work. The project checks all of the usual Kickstarter boxes. It’s open source and runs a version of Linux. The hardware is modular and ready for add-ons. SeeedStudio also says it’s building a bunch of APIs and an open SDK for developers to make cool stuff with ReSpeaker.
Hardware features like a far-field voice capture microphone array, acoustic echo cancellation, and beamforming, mean the ReSpeaker might actually be able to hear you as well as more expensive hardware like the Echo. ReSpeaker also has a standard audio out port to turn any speaker into a "smart" speaker.
The even better news is that ReSpeaker actually uses existing voice recognition services from the major players, like Amazon’s Alexa Voice Service, the Google Speech API, and the Microsoft Cognitive Service. All of these put together give the ReSpeaker a decent chance of not being as bad as, say, a DIY Amazon Echo built with a Raspberry Pi, cheap USB microphone, and a couple Python scripts.
Demos on the Kickstarer page include turning any speaker into an AirPlay-enabled talking speaker, a "what’s the weather?" type widget, and a plant that tells you when it needs water using data from SeeedStudio’s Wio line of IoT sensors.
ReSpeaker plans to ship in November for $69, but you can preorder it now on Kickstarter and get in on some sweet early-bird deals.