Proof of concept for voice commands on TFGrid/3bot/...

This was tested on python 3.11, though any python version since 3.8 should suffice. Installation platform was Arch linux, other platforms should work though at this stage YMMV.

Start by cloning the repo:

git clone https://github.com/LeeSmet/vcr
cd vcr

You probably want to use a virtual env, though it is not required. If you do want one:

python3 -m venv env
source ./env/bin/activate

Make sure you have the portaudio package installed (this is needed to capture the microphone input).

Ubuntu:

sudo apt-get install python-pyaudio python3-pyaudio

OSX:

brew install portaudio

For other platforms, check the PyAudio website

Now install the python packages from the requirements.txt file

pip install -r requirements.txt

After this, you should be good to go

Running the code is done by

python3 vcr/__main__.py

in the root of the repository. Alternatively the --cli flag can be used, in which case the program will run in a chat like mode