Improving user interaction with voice synthesis

Even if the Chronotherm application is working correctly, we have at least one more thing to do: providing proper feedback to let users know the action that was taken. Indeed, both activities fail to provide any visual feedback about what the recognized input is; for this reason, we decided to introduce the voice synthesis API within the initial design.

Because we want to share the synthesis procedure across different activities, we could create a manager that abstracts the synthesis API with a common initialization. The idea is to provide a class that exposes the method to start voice recognition with the given string; we implement it in the following steps:

1. Create the VoiceManager class inside the voice package.

2. Initialize the class with the following code:

   public class VoiceManager implements TextToSpeech.OnInitListener {
     private TextToSpeech mTts;
     //...
   }

   This class implements the OnInitListener interface that defines the callback that should...