Embedded speech for Arduino is here!

April 07, 2016

Max The Magnificent-April 07, 2016

Do you recall that MOVI (My Own Voice Interface) Kickstarter project from the summer of last year? I was pretty excited at the time (see Speech Recognizer / Synthesizer Shield for Arduino & Other MCUs).

In fact, I was so enthused by all of this that I managed to persuade MOVI's creators, Bertrand Irissou and Gerald Friedland, to give a presentation on the underlying technology at ESC Silicon Valley 2015 (see Learn How to Implement Embedded Speech at ESC Silicon Valley).


Bertrand Irissou (left) and Gerald Friedland (right) (Source: Audeme)

Well, time has been racing by as is its wont. I just heard that, following Bertrand and Gerald's very successful Kickstarter campaign, their company, Audeme, is powering along and MOVI is now available for purchase to makers and users worldwide!

As a brief summary:

  • MOVI is an easy to use speech recognizer/voice synthesizer for an Arduino shield.
  • MOVI can be easily programed to recognize 150 full-sentence voice commands.
  • MOVI is speaker independent and can carry on programmed conversations.
  • MOVI is privacy-friendly, non-cloud-based, with no Internet or PC connection required.

This type of technology has been long-awaited by makers because the ability to support voice commands and other auditory functions into their projects offers endless possibilities.


The MOVI speech recognition shield for Arduino (Source: Audeme)

I know that I for one am interested in adding this capability to my own hobby projects, including my BADASS Display, my Cunning Chronograph, my Nixie Tube Clock, and -- of course -- my Caveman Diorama.

Some possibilities are pretty obvious, like using voice commands to set the time on a clock, for example, but other possibilities are less obvious. In reality, you are limited only by your own imagination. In the case of the Caveman Diorama, for example, I was already planning on having sound effects like the chatter of conversation and music playing from a boom box (remember that the diorama will include a Time Portal, thereby explaining anachronisms like myself and my boom box being in the scene).

By means of MOVI, I can imagine myself saying things like "Hey, keep it down in there," and then having the diorama respond with a short period of silence followed by some angry grunting from the cavemen. Or I might say "That music sounds great, can you turn it up a bit?" The diorama could then respond with some accommodating grunts followed by an increase in the music's volume.

The more I think about this the more excited I get. The great news is that I just received an email from Bertrand and Gerald saying that they have a MOVI winging its way to me as we speak. If it arrives before next Tuesday, I'll bring it with me to show everyone at ESC Boston 2016. Hopefully I'll see you there.

Loading comments...