Detect simple voice commands

I would like to detect simple words or phrases from my microphone and perform actions based on those phrases. I've looked into Python libraries and Google text-to-speech but these seem like extreme overkill1. I don't need something that is capable of recognizing every phoneme or word in the English language, I just want to detect certain phrases like "go to sleep" or even just "sleep" to make my computer sleep for example. I tried searching for this, but mostly I just get programs for dictation and posts from 10 years ago. 1\. For example I stumbled across [this article](https://realpython.com/python-speech-recognition/) that relies on web services or installing something heavy-duty like Sphinx. Can't I just train a model to respond to certain phrases instead of every possible phrase?