Open a-t-0 opened 5 years ago
It is probably faster to copy an opensource voice recognition system than to build one.
There are systems that convert voice recognition into functions, look into it, one runs on raspberry pi, forgot the name, thought it was nips. Re-ask the CS TA working with it.
It saves time to be able to talk taskwarrior app. E.g. Android: taskwarrior filter project:U-N-I-DOT-S-S-E-space-"some description"
Android returns: task 5, description:"some discription of a random task that contains your keywords as an example"
User says: Android: Taskwarrior start task 5.
Android returns: Starting task 5.