Inspired by Lenny

Features:

Run a caller through a static voice script that you can easily change using text to speech.

Add in optional AI conversational twists using great (but pay for) Google speech APIs (see do_stuff_based_on_transcription in AsteriskTheSpammers.py). 

Hat tips:

https://www.youtube.com/results?search_query=lenny+asterisk

https://crosstalksolutions.com/howto-pwn-telemarketers-with-lenny/

https://raspberrytips.com/install-asterisk-on-raspberry-pi/

https://stackoverflow.com/questions/42439846/how-to-allow-inbound-calls-in-pjsip-and-asterisk-13

Installing
----------

-> sudo apt install these:

asterisk
tshark
dnsutils

-> sudo python3 -m pip install those:

pyst2
google-cloud-texttospeech 
google-cloud-speech
numpy

-> Place your (optional) google cloud speech api 'credentials.json' file in the location 
	indicted by <ConfigFiles/example> if you want AI stuff.

-> Put in your SIP provider details into <ConfigFiles/example>

my_incoming_id = 0123456789
my_local_network_address = 192.168.1.0/24
my_client_uri = sip:[email protected]:5060
my_server_uri = sip:sipgate.co.uk:5060
my_username = 0123456789
my_password = potatocrayfishbake
my_from_user = 0123456789
my_from_domain = sipgate.co.uk 
my_ethernet_interface = eth0

-> python3 ./configure.py ConfigFiles/your-config-file  

-> cd Asterisk/deploy  

-> sudo bash ./deploy.sh

For some reason, I could not get EAGI to work for access to incoming audio, hence the use of Monitor() 
	in the dialplan, and lock_onto_incoming_audio_file() in AsteriskTheSpammers.py. 
		Just found out about that https://github.com/CyCoreSystems/audiosocket though!