Voice Recognition Assisted Living


Developing my own household assistants controlled via voice or video recognition


Starting this project small scale at a time, the first step was to test feasibility of control using an already existing voice recognition software and having my voice commands control a simple motor, which would mimic as an automatic door

Implementation & Prototyping:

The motor and the hardware design was done with solidworks to build an attachment to the motor in order to increase surface area contact with the door. To interface and control the motor, i used an arduino.
I programmed the arduino using python, which allowed me to write my voice recognition algorithm and import libraries to control the arduino without having to use other software.

For prototpying purposes, I used AWS transcribe, which would listen to voice commands and return the data via an S3 bucket as text.

From there I would be able to make a request to the S3 bucket to retrieve the plain text of the voice command, and then use python to determine if the command was telling the motor to open the door or close it.
Then it would send the command to the arduino and motor to move it.

Future Work:

This is merely the first step in voice assisted living services.

In order to decrease response time, future steps would include using simple algorithm software to detect and return simple voice commands directly in python which would allow for no internet usage of the hardware.
In addition, being able to filter background noise would be beneficial to increasing accuracy of the system.

This project is a work in progress. Updates will follow.

Technology used:

Python, Eclipse, Arduino, Solidworks, AWS S3, Transcribe, Selenium


Additional Projects