服务中的文本转语音

Question

我在 Service 中遇到 TTS 问题。它表现得好像想要说话，但它从来没有说话。观察 LogCat，它会打印“TTS 已收到：它应该说出的文本”，并且我会在初始化时进行记录，这表明成功。我尝试为其创建一个线程，但这没有帮助。 onUtteranceComplete 也永远不会触发。我什至做了一个像这样的 while 循环（仅用于测试）：

while(mTTS.isSpeaking()) {
      Log.d("", "speaking");
}

...并且它从不说话

我知道 TTS 设置正确，因为它在常规 Activity 中工作，

这是我的代码。

import java.util.HashMap;
import java.util.Locale;

import android.app.Service;
import android.content.Intent;
import android.media.AudioManager;
import android.os.IBinder;
import android.speech.tts.TextToSpeech;
import android.speech.tts.TextToSpeech.OnInitListener;
import android.speech.tts.TextToSpeech.OnUtteranceCompletedListener;
import android.util.Log;

public class TTSService extends Service implements OnInitListener, OnUtteranceCompletedListener {
TextToSpeech mTTS;

    @Override
    public void onCreate() {
        Log.d("", "TTSService Created!");
        mTTS = new TextToSpeech(getApplicationContext(), this);


        //I've tried it in a thread....
        /*new Thread(new Runnable() {
            @Override
            public void run() {
                HashMap<String, String> myHashStream = new HashMap<String, String>();
                myHashStream.put(TextToSpeech.Engine.KEY_PARAM_STREAM, String.valueOf(AudioManager.STREAM_NOTIFICATION));
                myHashStream.put(TextToSpeech.Engine.KEY_PARAM_UTTERANCE_ID, "1");

                mTTS.setLanguage(Locale.US);
                //mTTS.setOnUtteranceCompletedListener(this);
                mTTS.speak("I'm saying some stuff to you!", TextToSpeech.QUEUE_FLUSH, myHashStream);        
            }

        }).start();*/

        //I've tried it not in a thread...
        HashMap<String, String> myHashStream = new HashMap<String, String>();
        myHashStream.put(TextToSpeech.Engine.KEY_PARAM_STREAM, String.valueOf(AudioManager.STREAM_NOTIFICATION));
        myHashStream.put(TextToSpeech.Engine.KEY_PARAM_UTTERANCE_ID, "1");

        mTTS.setLanguage(Locale.US);
        mTTS.setOnUtteranceCompletedListener(this);
        mTTS.speak("I'm saying some stuff to you!", TextToSpeech.QUEUE_FLUSH, myHashStream);    

    }

    @Override
    public IBinder onBind(Intent intent) {

        return null;
    }

    @Override
    public void onInit(int status) {
        Log.d("", "TTSService onInit: " + String.valueOf(status));
        if(status == TextToSpeech.SUCCESS){
            Log.d("", "TTS Success");
        }
    }

    public void onUtteranceCompleted(String uttId) {
        Log.d("", "done uttering");
        if(uttId == "1") {
            mTTS.shutdown();
        }

    }

}

谢谢

Answer 1

好的，我现在已经弄清楚了！发生的情况是它在初始化 TTS 之前尝试说话。因此，在一个线程中，我等待准备好不 == 999。一旦它为 1 或其他任何值，我们将负责说话。将其放入 while 循环中可能不安全，但是......它仍然有效。

import java.util.HashMap;
import java.util.Locale;

import android.app.Service;
import android.content.Intent;
import android.media.AudioManager;
import android.os.IBinder;
import android.speech.tts.TextToSpeech;
import android.speech.tts.TextToSpeech.OnInitListener;
import android.speech.tts.TextToSpeech.OnUtteranceCompletedListener;
import android.util.Log;

public class TTSService extends Service implements OnInitListener, OnUtteranceCompletedListener {
TextToSpeech mTTS;
int ready = 999;
    @Override
    public void onCreate() {

        Log.d("", "TTSService Created!");
        mTTS = new TextToSpeech(getApplicationContext(), this);

        new Thread(new Runnable() {
            @Override
            public void run() {
                while(ready == 999) {
                    //wait
                }
                if(ready==1){
                HashMap<String, String> myHashStream = new HashMap<String, String>();
                myHashStream.put(TextToSpeech.Engine.KEY_PARAM_STREAM, String.valueOf(AudioManager.STREAM_NOTIFICATION));
                myHashStream.put(TextToSpeech.Engine.KEY_PARAM_UTTERANCE_ID, "1");

                mTTS.setLanguage(Locale.US);
                //mTTS.setOnUtteranceCompletedListener(this);
                mTTS.speak("I'm saying some stuff to you!", TextToSpeech.QUEUE_FLUSH, myHashStream);

                } else { 
                    Log.d("", "not ready"); 
                    }
            }

        }).start();


        stopSelf();

    }

    @Override
    public IBinder onBind(Intent intent) {

        return null;
    }

    @Override
    public void onDestroy() {
        mTTS.shutdown();
        super.onDestroy();
    }
    @Override
    public void onInit(int status) {
        Log.d("", "TTSService onInit: " + String.valueOf(status));
        if (status == TextToSpeech.SUCCESS)
        {
            ready = 1;

        } else {
            ready = 0;
            Log.d("", "failed to initialize");
        }

    }

    public void onUtteranceCompleted(String uttId) {
        Log.d("", "done uttering");
        if(uttId == "1") {
            mTTS.shutdown();
        }

    }


}

Answer 2

我刚刚遇到了类似的问题 - 这是由于在 onInit() 启动 TTS 之前调用了 mTTS.speak。我的处理方式略有不同，将 mTTS.speak() 放在一个单独的函数中，然后从 onInit if (TextToSpeech.Successful) 中调用该函数，因此播放语音 <启动之后并且启动之前不（来自onStart()等）。