The past few years have seen a huge surge in the number of consumer devices with voice control features that have hit the market.
First came the voice assisting smartphone, with Google Assistant and Siri, and now making their way into the rest of your house with always-listening devices like Amazon Echo and starting this week, Google Home.
It’s an exciting time for device enthusiasts, as speech-to-text technology is finally getting to the point where it can be expected to be reliable and accurate without weirdly enunciating your voice. And with Amazon Echo and Google Home both seeing reduced prices for Black Friday sales this week, it’s likely that you or someone you know will end up with one of these devices at some point in the holiday season.
But where does open source fit into the picture? Is voice-controlled, connected future destined to be forever dominated by a few proprietary choices of custom-built hardware/software combinations that are essentially black boxes to their users? We hope not!
In fact, there are a few open source tools for voice control out there already, and it wouldn’t surprise me if the field grows as the technology becomes more pervasive. Looking for a weekend project? Check out a few of these options.
Alexa Skills Kit
The first stop on your journey might be in seeing how you can use open source tools to extend the functionality of your Amazon Echo. The Alexa Skills Kit lets you build custom commands for Alexa, the voice technology powering the Echo, and many of the example applications and tools are open source under an Apache license, including the Skills Kit SDK for Node.js. Be sure to check the licenses on the individual components, though. This might be a good first step if you’re hoping to integrate a newly acquired Echo with an open source home automation system.
While the Echo may be popular, there are completely open source alternatives. One such choice is Mycroft, a set of open source projects combining speech-to-text, text-to-speech, natural language processing, and intent processing (essentially, determining the meaning behind your request). After a successful crowdfunding campaign, Mycroft is continuing to develop towards a hardware product release, but the software development is happening in the open on GitHub, and is usable in your own projects if you’re willing to put in a little work; their dev kit runs a Raspberry Pi 3 under the hood.
Mycroft isn’t the only open source option for playing with voice control; another option is Jasper. Jasper is a voice control system designed for use with low-cost hardware, including the Raspberry Pi and other similar boards. It’s designed to be easily configurable and to provide an easy API for writing your own apps.
This only scratches the surface of some of the many ways that one can explore voice control with open source tools. I’m sure there are some great things you’ve run across out there that I’ve never even heard of! Share your favorite open source voice tool or open source integration with an Echo, Google Assistant, or other proprietary tool and let us all know some of the other cool things that are out there.