Real-time inference example

[elaith9]

Hello! I'm wondering if is possible to use those models in real-time inference scenario like microphone stream. In other words, will models work good on smaller chunks of audio like 250ms instead on whole sentences like those used for training. And if you have some example would you be able to provide it. Thanks, and great work!

[nglehuy]

@elaith9 Ideally the rnn_transducer can work on the small chunk of audio, but we need to test more 😄