- this is a python/django backend project. you'll need to have python 3.10ish (3.11 is fine), install Django & other requirements. and also have ngrok installed and working.
- you will also! need access to 1) a twilio phone number & api key 2) an openai account/api key (to do the speech recognition)
- roughly follow instructions from here
- fill out the missing values in the
.envfile (copied over from.env.example) i.e. the twilio & openai API keys - ping us in the issues here if you have trouble with any of this
source venv/bin/activatepython manage.py runserverngrok http 8000- then! go to twilio console for the phone number you're using and set:
- the dev number's voice's webhook URL...
- AND "call status changes" URL...
- ... to
https://...THE-NGROK-DOMAIN..../twilio_webhook