I want to build an app based on Python and a web UI 

The project goal is to build an all in one solution for tracking snooring and tracking sleep.

to do this we need to allow the end user to 
-  select an audio input device from the ones connected to the 'server'
-  to record ambiant room noise and gather samples of snoring
  -  to do this we'll need to define good sample sizes and periods of recording
  -  we'll also need to save the files in a audio format that doesn't pin the CPU/gpu but also doesn't waste filesize pointessly.
  -  We will likely also want to store the produced data from the sample that will be used by the network for training (i am thinking a basic spectrogram stored as a png)
  -  User will record samples over a couple nights, find examples of snoring, train the AI, let it try flagging the snoring events itself and once it's done trigger the anti-snoring events
-  must be self-hosted and self-trained, designed to be safe and appless so people are comfortable with passive recording