Arduino is a great platform for those wanting to get into both software and hardware development, but incorporating voice commands into these projects is not without its challenges. MOVI is a standalone speech recognition and synthesizer unit with full sentence capability that aims to make this task straightforward.
MOVI, which stands for "My Own Voice Interface," is interesting for a number of reasons. While most voice recognition systems built into consumer devices use cloud-based databases to understand what the user is saying, MOVI has a 2GB on-board database of up to 200 customizable sentences. Audeme, the company behind MOVI, says it is very easy to program and as an off-grid solution, it also ensures user privacy.
The MOVI Arduino shield is also a speech synthesizer, so developers can build conversations around their platforms and inventions. Audeme says that the device will respond in the same way to the same sentence every time, no matter who is talking to it. Not only that, but the device has a programmable call sign, meaning that users can personalize how to refer to the device according to their project.
The example given on the company's Kickstarter page for the use of the device is simple yet effective, showing a voice-controlled light switch, in which users can not only tell the device to switch the light on or off, but can also tell it to change color, or say things like "I’m feeling romantic" or "I want to relax," at which point the device will adjust the lamp accordingly.
Funds raised from the crowdfunding campaign will be used to merge the prototype, which is made out of a number of boards, into one single shield, after which it will be made available in the Maker community.
The MOVI Kickstarter campaign is a whisker away from reaching its $12K goal at the time of writing. A pledge of $80 puts you in line for a Movi shield and if all goes to plan, delivery is expected in early 2016.
The MOVI pitch video is below.