Let's Make Robots!

Audio - Process - Audio??

Audio - Process - Audio??

Guys I want to make a robot that can talk but before that I want to know how can voice be converted to text and then processed by some programming language and then the output is in the form is also audio?

For example if I say "5 + 2 = " I want a response "seven" (in audio)

I hope you guys understood what I wanted to tell.

Kindly help me with this, I will surely be grateful