See more
The first step in speech recognition is obvious — we need to feed sound waves into a computer.
To build a voice recognition system that performs on the level of Siri, Google Now!, or Alexa, you will need a lot of training data — far more data than you can likely get without hiring hundreds of people to record it for…