- when an incoming call arrives, the caller s audio is taken in real time and processed through an llm based on questions set by the user
- for example, a conversation may go like this
- caller: hi I am angel from XYZ spam company
- you: (autoresponder) who do you want to talk to?
- caller: ABC
- you: (autoresponder) validates ABC, if success next step else drop call
- you: (autoresponder) what is the password to talk to ABC?
- caller: I dont know
- you: (autoresponder) CALL DROPPED
- the user basically can set questions that the caller should answer and if the criteria is successful the call is accepted and the phone actually rings otherwise call is dropped
- how do I implement this on Android / iOS?