If you want more control over the speech recognition interaction, you can use the WebSockets interface for the Speech to Text service. One of the great benefits of the WebSockets interface is that it will give you interim results as the speech service is evaluating what is said. Besides giving you fascinating insights into how the speech engine works, its real-time feedback can be useful for helping users refine their own enunciation to help the engine do a better job of understanding the user.
Moreover, when you combine the WebSockets-based interface for speech recognition with streaming directly from a microphone on your machine, you can create a very speedy and responsive application ...