Using a simple speech recognition model in Android

The TensorFlow Android example app for simple speech commands recognition, located at tensorflow/example/android, has code that does audio recording and recognition in the SpeechActivity.java file, which assumes the app needs to be always ready for new audio commands. While this certainly makes sense in some cases, it also results in code that's more complicated than the code that would do recording and recognition only after the user presses a button, like the way Apple's Siri works. In this section, we'll show you how to create a new Android app and add the minimum possible code to record users' speech commands and display recognition results. This should help you more easily integrate the model to your own Android app. But if you need to deal with the case where speech commands should always be automatically...