依托于百度开放云，百度语音为合作伙伴提供了业界领先、永久免费的语音技术服务，目前已上线的服务包括语音识别、语义解析、语音合成，后续会继续带来开放资源、多轮对话等技术服务；通过SDK、REST API及离线开发包等多种服务方式，满足不同开发者的开发需求。

目前百度语音提供免费的语音接入，语音合成质量尚可，但还没开放离线TTS。

具体详情可查看：http://yuyin.baidu.com/dev.php

导入百度语音的相关jar包和so文件如图：

提示：

将开发包中的 libs目录整体拷贝到工程目录，libs目录包括了各平台的 SO库，开发者视应用需要可以进行删减。galaxy_lite.jar是百度 Android公共基础库，如果项目中还集成了其它百度 SDK，如 PushSDK，在打包过程中出现类似如下的错误信息：

[2013-10-22 11:02:57 - Dex Loader] Unableto execute dex: Multiple dex files defineLcom/baidu/android/common/logging/Configuration; [2013-10-22 11:02:57 -VoiceRecognitionDemo] Conversion to Dalvik format failed: Unable to executedex: Multiple dex files define Lcom/baidu/android/common/logging/Configuration;

请将此 Jar 包移除。如果 Eclipse ADT版本插件低于 17，需要手工添加依赖库，添加方法为：Project => Properties => Java Build Path => Libraries => AddJAR... 3

为了方便使用我这里将百度语音合成引擎做了进一步的封装，封装成了SpeechUtil以方便调用：

package com.jph.tts;import android.content.Context;
import android.media.AudioManager;
import android.util.Log;import com.baidu.speechsynthesizer.SpeechSynthesizer;
import com.baidu.speechsynthesizer.SpeechSynthesizerListener;
import com.baidu.speechsynthesizer.publicutility.SpeechError;/*** 语音合成工具类* * @author JPH* @date 2015-4-14 下午1:33:37*/
public class SpeechUtil implements SpeechSynthesizerListener {protected static final int UI_LOG_TO_VIEW = 0;private SpeechSynthesizer speechSynthesizer;private Context context;public SpeechUtil(Context activity) {this.context = activity;init();}/*** 初始化合成相关组件* * @author JPH* @date 2015-4-14 下午1:36:53*/private void init() {speechSynthesizer = new SpeechSynthesizer(context, "holder", this);// 此处需要将setApiKey方法的两个参数替换为你在百度开发者中心注册应用所得到的apiKey和secretKeyspeechSynthesizer.setApiKey("your apiKey","your secretKey");speechSynthesizer.setAudioStreamType(AudioManager.STREAM_MUSIC);
//      activity.setVolumeControlStream(AudioManager.STREAM_MUSIC);setParams();}/*** 开始文本合成并朗读* @author JPH* @date 2015-4-14 下午1:47:05* @param content*/public void speak(final String content) {new Thread(new Runnable() {@Overridepublic void run() {
//                 setParams();int ret = speechSynthesizer.speak(content.toString());if (ret != 0) {Log.e("inf","开始合成器失败："+ret);}}}).start();}/*** 取消本次合成并停止朗读* @author JPH* @date 2015-4-14 下午2:20:33*/public void cancle() {speechSynthesizer.cancel();}/*** 暂停文本朗读，如果没有调用speak(String)方法或者合成器初始化失败，该方法将无任何效果* @author JPH* @date 2015-4-14 下午2:21:07*/public void pause() {speechSynthesizer.pause();}/*** 继续文本朗读，如果没有调用speak(String)方法或者合成器初始化失败，该方法将无任何效果* @author JPH* @date 2015-4-14 下午2:21:29*/public void resume() {speechSynthesizer.resume();}/*** 为语音合成器设置相关参数* @author JPH* @date 2015-4-14 下午1:45:11*/private void setParams() {speechSynthesizer.setParam(SpeechSynthesizer.PARAM_SPEAKER, "0");//发音人，目前支持女声(0)和男声(1) speechSynthesizer.setParam(SpeechSynthesizer.PARAM_VOLUME, "9");//音量，取值范围[0, 9]，数值越大，音量越大speechSynthesizer.setParam(SpeechSynthesizer.PARAM_SPEED, "5");//朗读语速，取值范围[0, 9]，数值越大，语速越快speechSynthesizer.setParam(SpeechSynthesizer.PARAM_PITCH, "5");//音调，取值范围[0, 9]，数值越大，音量越高speechSynthesizer.setParam(SpeechSynthesizer.PARAM_AUDIO_ENCODE,SpeechSynthesizer.AUDIO_ENCODE_AMR);//音频格式，支持bv/amr/opus/mp3，取值详见随后常量声明speechSynthesizer.setParam(SpeechSynthesizer.PARAM_AUDIO_RATE,SpeechSynthesizer.AUDIO_BITRATE_AMR_15K85);//音频比特率，各音频格式支持的比特率详见随后常量声明}@Overridepublic void onStartWorking(SpeechSynthesizer synthesizer) {Log.i("msg", "开始工作，请等待数据...");}@Overridepublic void onSpeechStart(SpeechSynthesizer synthesizer) {Log.i("msg", "朗读开始");}@Overridepublic void onSpeechResume(SpeechSynthesizer synthesizer) {Log.i("msg", "朗读继续");}@Overridepublic void onSpeechProgressChanged(SpeechSynthesizer synthesizer,int progress) {// TODO Auto-generated method stub}@Overridepublic void onSpeechPause(SpeechSynthesizer synthesizer) {Log.i("msg", "朗读已暂停");}@Overridepublic void onSpeechFinish(SpeechSynthesizer synthesizer) {Log.i("msg", "朗读已停止");}@Overridepublic void onNewDataArrive(SpeechSynthesizer synthesizer,byte[] audioData, boolean isLastData) {Log.i("msg", "新的音频数据：" + audioData.length+ (isLastData ? "(end)" : ""));}@Overridepublic void onError(SpeechSynthesizer synthesizer, SpeechError error) {Log.i("msg", "发生错误：" + error.errorDescription + "(" + error.errorCode+ ")");}@Overridepublic void onCancel(SpeechSynthesizer synthesizer) {Log.i("msg", "已取消");}@Overridepublic void onBufferProgressChanged(SpeechSynthesizer synthesizer,int progress) {// TODO Auto-generated method stub}
}

使用方法：

package com.jph.tts;import android.app.Activity;
import android.os.Bundle;
import android.text.method.ScrollingMovementMethod;
import android.view.View;
import android.view.View.OnClickListener;
import android.widget.Button;
import android.widget.EditText;
import android.widget.TextView;import com.baidu.tts.sample.R;public class HomeActivity extends Activity implements OnClickListener {protected static final int UI_LOG_TO_VIEW = 0;private TextView logView;private EditText inputTextView;private Button startButton;private SpeechUtil speechUtil;@Overrideprotected void onCreate(Bundle savedInstanceState) {super.onCreate(savedInstanceState);setContentView(R.layout.activity_home);logView = (TextView) findViewById(R.id.logView);logView.setMovementMethod(new ScrollingMovementMethod());inputTextView = (EditText) findViewById(R.id.inputTextView);speechUtil = new SpeechUtil(this);}@Overridepublic void onClick(View v) {switch (v.getId()) {case R.id.btnStart:speechUtil.speak(inputTextView.getText().toString());break;case R.id.btnPause:speechUtil.pause();break;case R.id.btnResume:speechUtil.resume();break;case R.id.btnStop:speechUtil.cancle();break;default:break;}}
}

因为百度TTS需要联网解析所以要赋予应用相应的权限：

<uses-permission android:name="android.permission.INTERNET"></uses-permission><uses-permission android:name="android.permission.ACCESS_NETWORK_STATE"></uses-permission><uses-permission android:name="android.permission.READ_PHONE_STATE"/>

推荐阅读：

【Android语音合成TTS】国内主流引擎对比 
【Android语音合成TTS】云知声离线TTS使用详解