forecast-gen
A simple python system to generate products similar to NOAA Weather Radio and others.
How to use
Edit the variables listed under the setup comment in start_system.py.
(Forecast zones & observation stations can be found here: https://api.weather.gov/offices/your_office_here)
Once those are all set up, simply run start_system.py in python. This will generate a fully-formatted text product in speech.txt.
If needed, you can edit some of the functions in main.py to suit what you want to output into the file in terms of formatting.
TTS Setup
Any forecast-gen version starting from 0.2.0 supports TTS functionality.
Software
Forecast-gen uses a software called Balcon to read off the text file generated from running start_system.py. As for voices, any SAPI 5 voice should work perfectly fine with the software above. If you're looking for a voice to start off with, you could try the voices from the MS TTS Pack, or the SAPI 5 voices that come included in Windows 10.
Running SAPI 5 voices through WINE (Linux, macOS)
Note: Most of these instructions are for Linux, as I don't use macOS
Under the assumption that you have already installed WINE and winetricks through your package manager of choice, create a 32-bit wineprefix so you can run Balcon, and then run winetrick speechsdk in your terminal. Once those are done, run wine balcon -l in the directory where Balcon is in, and it should list off some voices, by default speechsdk installs Microsoft Sam, Mike, and Mary.
Notes
The software for TTS support on Linux, as well as the software for using the voices in a Linux terminal, will not work on ARM operating systems such as Raspbian or Arch Linux ARM.
[22/05/2019] I've decided that I might eventually remake the CRS system at some point, whether it be in python or some other language. This will come with DECTalk as a standalone voice, but if you have the right voices (RealSpeak Tom, AT&T Crystal & Mike) those can be enabled as well.