Digital assistant apps of all kinds are on the rise, but you don’t necessarily need to invest in a smart speaker to start chatting to one: with a little know-how you can get Amazon Alexa or Google Assistant up and running on your computer. Here’s what you need to do.
If you suddenly find yourself taken with the idea of chatting to bots on your laptop, then remember you can fire up either Cortana on Windows or Siri on macOS for a little conversation. What’s more, it’s likely only a matter of time before Google’s digital assistant comes to Chrome and Chrome OS, but it’s not there yet.
Reasons to get chatty
Firing up Alexa on an Echo or the Google Assistant on Google Home (or an iPhone) are easier routes for anyone wanting to get answers, calendar update,s and restaurant recommendations with their voice. It requires a few minutes of hassle to get the assistants running on a computer, but that doesn’t mean it’s not worth it.
Maybe you’ve got an Amazon Echo installed downstairs but you’ve only got your laptop upstairs; or maybe you’re away from home altogether and pining for some interaction with Alexa. There are quite a few reasons why you would want to set this up and the desktop apps sync neatly with your existing accounts, so all your stuff is still available.
Our Google Assistant hack is a little more time-consuming, but again, it allows you to test out what Google believes is the future of interaction right from the standard laptop or desktop you already own—and then you can decide if it’s worth splurging on a Pixel. Oh, and you can maybe impress the geekiest of your friends, too.
Amazon Alexa on the desktop
Amazon has actually released a web interface for Alexa, which you can find here. Sign in with your Amazon credentials to access it, and if you own an Echo at home, then you’ll find the screen populated with all your information and recent searches. You can’t use this screen to interact with Alexa, though.
You can, however, interact with Echosim.io, a skills testing ground for the web built by Amazon. The web app is aimed at developers but available to anyone with an Amazon account. Once you login, accept the terms and conditions, and grant your web browser access to your microphone, you can start interacting.
Click and hold the microphone button or hold the Space bar on your keyboard to start chatting (no need to say “Alexa” here). As you’re using your Amazon account, everything you’ve set up with your Echo—from calendars to shopping lists—can be accessed, and all your queries go into your main Alexa history too.
Any of the skills you can use with an Echo you can use with Echosim.io, though, obviously some require extra setup in the Alexa app, so you can’t start ordering Ubers and pizzas straight off the bat. One or two skills we tried seemed to get confused that they were sitting in a web browser, but generally speaking everything runs smoothly.
Reverb.ai is a very similar tool which also runs in any desktop web browser. Again, you need to click and hold on the microphone button, once your browser has been given permission to access your computer’s microphone, but since the app isn’t linked to your Amazon account—you’re limited to general, non-personal queries. Also we couldn’t get any third-party skills working. The Reverb app for macOS does link to your Amazon account, however, so you can pull up your calendar, try out different skills, and so on, just as if you were talking to a physical Amazon Echo.
Google Assistant on the desktop
Getting Google Assistant on the desktop is a lot more involved. Note that you can get some assistant-style functionality, like answers to simple web queries and your Google Calendar schedule, from Google.com in your browser by just clicking on the microphone button by the search box and speaking your commands.
For the full Google Assistant experience, though, you have to follow these instructions from Mishaal Rahman on XDA Developers. It’s a pretty complex process, which needs a lot of carefully constructed command line instructions—and at the end of it all you can only access Google Assistant through another command line window, though speech input and output is supported.
The time and effort required is more than some people will want to invest just for the novelty of having Google Assistant running on their desktop computer, but if you enjoy tinkering around with this kind of tech, and need something to occupy your mind, it’s worth a go. The bot gets linked to your Google account, so there’s no problem pulling up personal details from Gmail, Google Calendar and so on.
There’s no point us simply copying out all of the steps verbatim here—just head to XDA Developers for detailed instructions—but the process involves setting up Python, authorizing new credentials for the Google Assistant API, and then setting up a simple client in the command line. If all that text looks a bit daunting, Hayato Huseman at 9to5Google has put together a video explainer for macOS that’s very handy.
We got the Assistant up and running on a macOS machine using the supplied instructions in about half an hour, though if you’re not familiar with the command line, and tools like Xcode, you might find it takes you longer. The process of getting Python installed is slightly more straightforward on Windows, incidentally, if you have a choice between OSes.
By the end you can run a simple command line prompt to launch Google Assistant, hitting Enter every time you want to say something new—responses are spoken back to you, so there’s no text to read on screen (though you can check that the Assistant parsed what you said correctly). It’s a rather awkward workaround, but it’s fun to try out until Google finds time to start rolling out its Assistant everywhere.