Full project: https://bit.ly/35YOgGm
The name Just a Rather Very Intelligent System (JARVIS) is from the movie Iron Man. In this movie JARVIS is used by Tony Stark as his intelligent voice assistant. JARVIS in this project is an Artificial Intelligence (AI) system that responds visually to a set of pre-trained audial keywords. With the advancement in the field of Machine Learning (ML), tiny devices such as microcontrollers are capable of running ML models. This field is now termed as tinyML. At the moment, there is a lot of cutting -edge research and development that is being done in this field. It is always exciting to participate in the latest tech-trend, is it not? This motivated me to build an application with the board provided for the NXP Hackathon 2021.
I came across the NXP Hackathon 2021 competion while browsing through the social media. At that time I was planning on experimenting with CANopen (Software stack). After taking a look at the two boards that were available as the base kits, I chose iMX.RT1020 as the development boards. The reason is that these boards have hardware capabilities to support a CAN (Controller Area Network). There are very few examples that demonstrate the use cases of CANopen stack. I thought it is a good idea to integrate CANopen with an Audio Classifier and build a solution around it.
KEYWORD | Action
BRIGHTER | Increase the brightness
DIMMER | Reduce the brightness
DOWN | Turn-on the bottom side LED’s
UP | Turn-on the top side LED’s
LEFT | Turn-on the left side LED’s
RIGHT | Turn-on the right side LED’s
ON | Turn-on all the LED’s
OFF | Turn-off all the LED’s
YES | Turn-on top-left Green LED
NO | Turn-on top-left Blue LED
GO | Turn-on the corner four Green LED.
STOP | Turn-on the corner four Blue LED.
BACKGROUND | No Action
Blink mode: Turn-on if the same command is spoken with a slight pause.