I have tried playing the TextToSpeech object when a specific event is triggered in the phone.
However, I facing issues with the default Google TTS engine that is installed on most phones. As of now, I am playing some text immediately after the TextToSpeech object is initialized, and shutting the resource as soon as the speech is completed, as per the following code:
public class VoiceGenerator {
private Context context = null;
private static TextToSpeech voice = null;
public VoiceGenerator(Context context)
{
this.context = context;
}
public void voiceInit(String text)
{
try {
if (voice == null) {
new Thread(new Runnable() {
@Override
public void run() {
voice = new TextToSpeech(context, new TextToSpeech.OnInitListener() {
@Override
public void onInit(final int status) {
try {
if (status != TextToSpeech.ERROR) {
voice.setLanguage(Locale.US);
Log.d("VoiceTTS", "TTS being initialized");
HashMap p = new HashMap<String, String>();
p.put(TextToSpeech.Engine.KEY_PARAM_UTTERANCE_ID, "ThisUtterance");
//Speaking here
voice.speak(text, TextToSpeech.QUEUE_ADD, p);
voice.setOnUtteranceProgressListener(new UtteranceProgressListener() {
@Override
public void onStart(String utteranceId) {
}
@Override
public void onDone(String utteranceId) {
Log.d("VoiceTTS", "TTS being released");
clearTtsEngine();
}
@Override
public void onError(String utteranceId) {
}
});
}
} catch (Exception e) {
clearTtsEngine();
Log.d("ErrorLog", "Error occurred while voice play");
e.printStackTrace();
}
}
});
}
}).start();
}
}
catch(Exception e)
{
clearTtsEngine();
Log.d("ErrorLog","Error occurred while voice play");
e.printStackTrace();
}
}
public static void clearTtsEngine()
{
if(voice!=null)
{
voice.stop();
voice.shutdown();
voice = null;
}
}
}
However, the problem I am facing is the finite amount of delay associated with initializing the Google TTS Engine - about 6-8 seconds on my devices.
I have read on other posts that this delay can be avoided by using other TTS engines. Since I always develop on my Samsung phone, which has its own proprietary TTS configured by default, I never noticed this issue until I checked my app on other brand phones which has the Google TTS engine configured as default. But, I ideally don't want to force users to install another app along with my own, and I hence I would like this to work with the default Google TTS Engine itself.
Through some erroneous coding which I later rectified, I realized that if I could keep the TextToSpeech object initialized beforehand and always not null - once initialized, I could seemingly bypass this delay.
However, since there is a necessity to shutdown the resource once we are done with it, I am not able to keep the object alive and initialized for long, and I do not know when to initialize/shutdown the resource, since I technically need the voice to play anytime the specific event occurs, which mostly would be when my app is not open on the phone.
So my questions are the following :
Can we somehow reduce or eliminate the initialization delay of Google TTS Engine, programmatically or otherwise?
Is there any way through which I can keep the TextToSpeech object alive and initialized at all times like say, through a service? Or would this be a bad, resource-consuming design?
Also is using a static TextToSpeech object the right way to go, for my requirements?
Any solutions along with code would be appreciated.
Update: I have confirmed that the delay is associated exclusively with Google TTS engine, as I have tried using other free and paid TTS engines, wherein there is little or no lag. But I would still prefer to not have any third party dependencies, if possible, and would like to make this work with Google TTS Engine.
UPDATE: I have seemingly bypassed this issue by binding this TTS object to a service and accessing it from the service. The service is STICKY (if the service terminates due to memory issue, Android OS will restart the service when memory is available again) and is configured to restart on reboot of the device.
The service only initializes the TTS object and does no other work. I am not explicitly stopping the service, allowing it to run as long as possible. I have defined the TTS object as a static, so that I can access it from other classes of my app.
Although this seems to be working amazingly well, I am concerned if this could lead to memory or battery issues (in my specific situation where service handles only object initialization and then remains dormant). Is there any problem in my design, or can any further improvements/checks be done for my design?
Manifest file :
<uses-permission android:name="android.permission.RECEIVE_BOOT_COMPLETED"/>
<application
android:allowBackup="false"
android:icon="@drawable/ic_launcher"
android:label="@string/app_name" >
<activity
android:name="activity.MainActivity"
android:label="@string/app_name"
android:screenOrientation="portrait" >
<intent-filter>
<action android:name="android.intent.action.MAIN" />
<category android:name="android.intent.category.LAUNCHER" />
</intent-filter>
</activity>
<receiver
android:name="services.BroadcastReceiverOnBootComplete"
android:enabled="true"
android:exported="false">
<intent-filter>
<action android:name="android.intent.action.BOOT_COMPLETED" />
</intent-filter>
<intent-filter>
<action android:name="android.intent.action.PACKAGE_REPLACED" />
<data android:scheme="package" />
</intent-filter>
<intent-filter>
<action android:name="android.intent.action.PACKAGE_ADDED" />
<data android:scheme="package" />
</intent-filter>
</receiver>
<service android:name="services.TTSService"></service>
BroadcastReceiver code :
public class BroadcastReceiverOnBootComplete extends BroadcastReceiver {
@Override
public void onReceive(Context context, Intent intent) {
if (intent.getAction().equalsIgnoreCase(Intent.ACTION_BOOT_COMPLETED)) {
Intent serviceIntent = new Intent(context, TTSService.class);
context.startService(serviceIntent);
}
}
}
TTSService code:
public class TTSService extends Service {
private static TextToSpeech voice =null;
public static TextToSpeech getVoice() {
return voice;
}
@Nullable
@Override
public IBinder onBind(Intent intent) {
// not supporting binding
return null;
}
public TTSService() {
}
@Override
public int onStartCommand(Intent intent, int flags, int startId) {
try{
Log.d("TTSService","Text-to-speech object initializing");
voice = new TextToSpeech(TTSService.this,new TextToSpeech.OnInitListener() {
@Override
public void onInit(final int status) {
Log.d("TTSService","Text-to-speech object initialization complete");
}
});
}
catch(Exception e){
e.printStackTrace();
}
return Service.START_STICKY;
}
@Override
public void onDestroy() {
clearTtsEngine();
super.onDestroy();
}
public static void clearTtsEngine()
{
if(voice!=null)
{
voice.stop();
voice.shutdown();
voice = null;
}
}
}
Modified VoiceGenerator code:
public class VoiceGenerator {
private TextToSpeech voice = null;
public VoiceGenerator(Context context)
{
this.context = context;
}
public void voiceInit(String text)
{
try {
if (voice == null) {
new Thread(new Runnable() {
@Override
public void run() {
voice = TTSService.getVoice();
if(voice==null)
return;
voice.setLanguage(Locale.US);
HashMap p = new HashMap<String, String>();
p.put(TextToSpeech.Engine.KEY_PARAM_UTTERANCE_ID, "ThisUtterance");
voice.speak(text, TextToSpeech.QUEUE_ADD, p);
voice.setOnUtteranceProgressListener(new UtteranceProgressListener() {
@Override
public void onStart(String utteranceId) {
}
@Override
public void onDone(String utteranceId) {
}
@Override
public void onError(String utteranceId) {
}
});
}
}).start();
}
}
catch(Exception e)
{
Log.d("ErrorLog","Error occurred while voice play");
e.printStackTrace();
}
}
}
