This repo gives a python package and CLI to interface with the Brillibot API.
Brillibot is an easy to use speech to command API.
It works by having a user simply define their possible commands in a JSON file with several ways they could say that command, and then provides either a file or access to a microphone to send audio over to sever that will determine what command was said in the audio.
The API is currently rate limited for public use, API keys are a possibility.
Redeployment server side takes a few minutes, thus due to the fact that development is still ongoing, expect periodic downtime.
ffmpeg is required for this application to work properly
It is reccomended to use a venv before installing.
pip install git+https://github.com/mallorbc/brillibot-client.git
If the repo was updated since you last installed, running may be needed:
pip install git+https://github.com/mallorbc/brillibot-client.git --no-cache
Inside the /examples folder there are two files that show how one can use the Python package either with a file, or with a microphone.
To use the CLI, insure that you are inside the venv after installation. The CLI is called brillibot_client
To see the run options run brillibot_client -h. Providing an actions file is a must, to do so, use the -a flag
This repo needs PortAudio to work.
If you are having issues with the your microphone, try running the following for Ubuntu:
sudo apt install portaudio19-dev python3-pyaudio gcc
Windows:
pip install pipwin
pipwin install pyaudio
This repo is licensed under the AGPL license. See LICENSE for more information. If you require other license options, contact me.