Alexa as Mycroft mic / speaker

I know Alexa doesn’t have the processing power to become a Mycroft by wondering if it could become the front-end mic/ speaker to Mycroft running on my Linux server

1 Like

Hey Arden, welcome to the Forums :slight_smile:

The short answer is unfortunately - it’s very unlikely.

I’ve seen a few people ask similar questions - more often about Google Mini’s, and I recently found a detailed write up of one persons attempts to get custom code running on them:

It’s a two part series and very detailed but also shows just how much effort is required to get any new code running on them.

So you are very welcome to try, but unless the challenge really excites you, I would look at some other more open hardware to replace it.

Anyone tried with other devices, just have one transmit audio to another, more robust “server” device that processes and sends the response?

Read the Alexa are underpowered, but you can get Debian on them. I know you can SSH into a Mycroft device, enter the cli, send a text entry, and receive a text response.

Can’t quite figure out how to send audio and have it “entered” in the same way, because the Alexa probably can’t handle the speech recognition work.

This could be a pipe dream, I just have too many of these stupid things and would love to use them somhehow rather than buying a new RPi 4 off a wait list. Will update if I have any luck.

I think this is still not possible. Only because I don’t see a way to get alexa to just pass the speech as text.

If there is a way to do that, then you should be able to bridge alexa and mycroft using nodered.

Once the OVOS Foundation (see FAQ) is incorporated, we intend to revive a dormant project that provides thin clients.

This would do basically what you ask. Your Amazon devices would become “satellite” nodes for one or more centrally-served assistants.

1 Like