freetalkbot

Your own IA to handle communications with your customers via phonecalls or whatsapp

Dependencies

Go version recommended: 1.22
whatsapp-media-decrypt tool
picotts
Install dependencies with go mod tidy

For better golang developer experience you can install golang-air

Environment variables

Check the variables in env.example file. Create .env file with cp -a .env.example .env and modify it with your values. All bot servers and Rasa assistant reads configuration from this file.

Development

The components added in the docker-compose.yml file are:

Asterisk
Rasa assistant
Rasa Actions server
Audio bot server
Whatsapp bot server

Build docker images

Run make build

Run the solution

Run make run

Configure asterisk

Once raised up, copy local-config to container-config cp -a asterisk/local-config/* asterisk/container-config/
Restart asterisk container docker-compose restart asterisk

Asterisk is raised up in network_mode brige. The asterisk configuration files are mapped in folder asterisk/container-config

Register SIP/IAX endpoint in a softphone

For SIP checkout pjsip_endpoint.conf file in asterisk/container-config folder.
For IAX checkout iax.conf file in asterisk/local-config folder.

Rasa assistant

The assistant is configured to be a reminderbot, inspired in the example provided by rasa. The files in this folder are for NLP training of the assistant. Checkout the following resources to get more knowledge about RASA.

If you modify any of the rasa files you will need to retrain the assistant, you can do it with make rasa-train

Disclaimer : this bot is not complete yet. It is doing what makes reminderbot example. I am still in the learning path of rasa and NLP.

Communication Channels

You can communicate with your chatbot assistant via two channels.

Audio channel

The docker-compose file alreay deploys it, but if you want to develop on it locally instead in the docker container:

Change envars in .env file:

RASA_URL=http://localhost:5005
CALLBACK_SERVER_URL=http://host.docker.internal:5034/bot

Set the envars with export $(cat ./.env | xargs)
Install pkg-config: brew install pkg-config
Change the audiosocket server host in asterisk/container-config/extensions_local.conf

 same = n,AudioSocket(40325ec2-5efd-4bd3-805f-53576e581d13,host.docker.internal:8080)

Run make run-local-audio

Whatsapp channel

Same variables than audio bot are needed, just change the make command make run-local-whatsapp

After initialize you will see in the logs a QR code. Scan that QR code with the whatsapp account that you will use. If you can't scan the QR code you can also link the whatsapp account using a pair code. For that you must set the envar PAIR_PHONE_NUMBER with your phone number using format show in the .env.example. If you don't need the pair code don't set this envar.

Once you pair your whatsapp account the session will be stored in a sqlite file. This file is created inside the container but mapped through a docker volume, so you can use it when you want to develop locally. If you delete this file you will have to login again using a new QR code.

Features

Now the code is prepared to receive text or voice messages. The assistant on this repository is a reminderbot that will send you reminders based on messages that you are sending to it

Gratitude and Thanks

The following projects inspired to the construction of this one:

nrjchnd/freetalkbot