This would be trivial to build with Amazon speech-to-text / text-to-speech (or IBM's offering for that matter)... I expect there is more than a few of us that h

This would be trivial to build with Amazon speech-to-text / text-to-speech (or IBM's offering for that matter)... I expect there is more than a few of us that have considered building this, so I'd be surprised if it didn't exist by now... although getting traction / monetisation is another thing to support costs...