一,jsp 网页使用JavaScript来实现文字转语音即TTS,TEXT TO speach, javascript
好比: php
仍是微软官方的方便简单,否则文字转语音在网页上播报这种功能真就麻烦了,特别是汉字转语音播报。这是使用js来实现的。 html
其次,经过Java实现中文到英文朗读出来 java
code: android
import java.awt.Container; import java.awt.Dimension; import java.awt.EventQueue; import java.awt.event.ActionEvent; import java.awt.event.ActionListener; import javax.swing.DefaultComboBoxModel; import javax.swing.JButton; import javax.swing.JComboBox; import javax.swing.JFrame; import javax.swing.JLabel; import javax.swing.JPanel; import javax.swing.JScrollPane; import javax.swing.JSeparator; import javax.swing.JSlider; import javax.swing.JTextArea; import javax.swing.UIManager; import org.loon.framework.jtts.Engine;
import org.loon.framework.jtts.JTTS; public class TalkerTest extends JPanel implements ActionListener { private static final long serialVersionUID = 1L; private JTTS jtts; private JButton jButton1; private JButton jButton2; private JComboBox jComboBox1; private JLabel jLabel1; private JLabel jLabel2; private JLabel jLabel3; private JLabel jLabel4; private JScrollPane jScrollPane1; private JSeparator jSeparator1; private JSlider jSlider1; private JSlider jSlider2; private JTextArea jTextArea1; @SuppressWarnings("unchecked") public TalkerTest() { this.jtts = Engine.getTTS(); this.jSlider1 = new JSlider(); this.jSlider2 = new JSlider(); this.jComboBox1 = new JComboBox(); this.jLabel1 = new JLabel(); this.jLabel2 = new JLabel(); this.jLabel3 = new JLabel(); this.jSeparator1 = new JSeparator(); this.jButton1 = new JButton(); this.jButton2 = new JButton(); this.jScrollPane1 = new JScrollPane(); this.jTextArea1 = new JTextArea(); this.jLabel4 = new JLabel(); setLayout(null); this.jSlider1.setName("jSlider1"); this.jSlider1.setMaximum(500); this.jSlider1.setValue(150); this.jSlider1.setMinimum(0); add(this.jSlider1); this.jSlider1.setBounds(94, 131, 316, 24);
this.jSlider2.setName("jSlider2"); this.jSlider2.setMaximum(200); this.jSlider2.setValue(100); this.jSlider2.setMinimum(0); add(this.jSlider2); this.jSlider2.setBounds(94, 170, 316, 24); String[] types = new String[12]; for (int i = 0; i < types.length; i++) { if (i < 5) types[i] = ("zh+m" + (i + 1)); else { types[i] = ("zh+f" + (i - 4)); } } types[10] = "zh+whisper";
types[11] = "zh+croak";
this.jtts.setLanguage("zh");
this.jComboBox1.setModel(new DefaultComboBoxModel(types));
this.jComboBox1.setName("jComboBox1"); add(this.jComboBox1);
this.jComboBox1.setBounds(100, 210, 100, 23);
this.jComboBox1.addActionListener(this);
this.jLabel1.setText("朗读间隔:");
this.jLabel1.setName("jLabel1");
add(this.jLabel1);
this.jLabel1.setBounds(20, 125, 66, 30);
this.jLabel2.setText("朗读音量:");
this.jLabel2.setName("jLabel2");
add(this.jLabel2);
this.jLabel2.setBounds(20, 165, 70, 30);
this.jLabel3.setText("发音规则:");
this.jLabel3.setName("jLabel3"); add(this.jLabel3);
this.jLabel3.setBounds(20, 216, 80, 15);
this.jLabel4.setText("目标文本:");
this.jLabel4.setName("jLabel4");
add(this.jLabel4);
this.jLabel4.setBounds(20, 10, 70, 20);
this.jSeparator1.setName("jSeparator1");
add(this.jSeparator1);
this.jSeparator1.setBounds(10, 268, 398, 10);
this.jButton1.setText("朗读文本");
this.jButton1.setName("jButton2");
add(this.jButton1);
this.jButton1.setBounds(232, 284, 83, 25);
this.jButton1.addActionListener(this);
this.jButton2.setText("退出应用");
this.jButton2.setName("jButton3");
add(this.jButton2);
this.jButton2.setBounds(325, 284, 83, 25);
this.jButton2.addActionListener(this);
this.jScrollPane1.setName("jScrollPane1");
this.jTextArea1.setColumns(20);
this.jTextArea1.setRows(5);
this.jTextArea1.setName("jTextArea1");
this.jTextArea1
.setText("Hello Java!Hello World!\n史无前例啊,中国踢韩国三比零啦!\n生意兴隆,财源广进,万事如意,虎年大吉!");
this.jScrollPane1.setViewportView(this.jTextArea1); add(this.jScrollPane1);
this.jScrollPane1.setBounds(98, 10, 310, 106);
} public void actionPerformed(ActionEvent e) { Object source = e.getSource(); if ((source instanceof JButton)) { if (source == this.jButton1) { this.jtts.setRate(this.jSlider1.getValue()); this.jtts.setVolume(this.jSlider2.getValue()); this.jtts.speak(this.jTextArea1.getText().trim()); } if (source == this.jButton2) System.exit(0); } else if ((source instanceof JComboBox)) {
this.jtts.setLanguage(this.jComboBox1.getSelectedItem().toString() .trim()); } } public static void createGUI() { try { UIManager.setLookAndFeel(UIManager.getSystemLookAndFeelClassName()); } catch (Exception localException) { }
JFrame frame = new JFrame("Java版TTS组件发音测试(eSpeak封装)"); frame.setDefaultCloseOperation(3); frame.getContentPane().add(new TalkerTest()); frame.pack(); frame.setSize(new Dimension(420, 345)); frame.setResizable(false); frame.setLocationRelativeTo(null); frame.setVisible(true); } public static void main(String[] args) { EventQueue.invokeLater(new Runnable() { public void run() { TalkerTest.createGUI();
} }); }}
图: web
二,经过 编程
语音合成是实现人机语音交互,跟机器说话,创建一个有听和讲能力的交互系统所必需的关键技术。随着语音技术的发展,百度自主研发了语音合成系统(TTS),功能是接受用户发送的文本,生成语音发送给用户。
对本文中将提到的概念约定以下:
语音合成:将文本合成为语音,即声音文件。
合成引擎:将文本合成为语音的核心模块。
TTS:Text To Speech,即“从文本到语音”。
BDTTSClient:语音合成 SDK 简称,详见下条。
语音合成 SDK:即本开发包,文中简称为 BDTTSClient。BDTTSClient 是一个封装了网络收发、音频播放功能的语音合成解决方案。借助 BDTTSClient 能够快速地在应用程序中集成语音合成功能。 json
1.下载Android 平台 HTML5+ SDK 。最好Android studio
2.注册百度开发者,建立应用,开通服务,请参考集成指南。
3.下载在线语音合成SDK_Android版。 windows
1.先导入5+SDK就很少说了,参考Android平台第三方插件开发指导。
2.添加BDTTSClient到工程(将开发包中的 libs 目录总体拷贝到工程目录,libs 目录包括了各平台的 SO 库,开发者视应用须要能够进行删减。galaxy_lite.jar 是百度 Android 公共基础库。)
3.权限声明(须要在 AndroidManifest.xml 文件,增长以上三个权限) api
<uses-permission android:name="android.permission.INTERNET" />
<uses-permission android:name="android.permission.ACCESS_NETWORK_STATE" />
<uses-permission android:name="android.permission.READ_PHONE_STATE" />
集成java代码
import org.json.JSONArray;
import com.baidu.speechsynthesizer.SpeechSynthesizer;
import com.baidu.speechsynthesizer.SpeechSynthesizerListener;
import com.baidu.speechsynthesizer.publicutility.SpeechError;
import android.media.AudioManager;
import android.util.Log;
import io.dcloud.common.DHInterface.IWebview;
import io.dcloud.common.DHInterface.StandardFeature;
import io.dcloud.common.util.JSUtil;
public class baidutts extends StandardFeature implements SpeechSynthesizerListener{
private IWebview iwv ;
private String text;
private SpeechSynthesizer speechSynthesizer;
private static final String TAG = "zlz";
public void speak(IWebview pWebview, JSONArray array)
{
iwv = pWebview;
String CallBackID = array.optString(0);
JSONArray newArray = new JSONArray();
newArray.put(array.optString(1));
//前台传过来的文本信息
text = array.optString(1);
//Log.i(TAG, text);
initialTts();
JSUtil.execCallback(pWebview, CallBackID, newArray, JSUtil.OK, false);
}
//百度TTS初始化
private void initialTts() {
speechSynthesizer = new SpeechSynthesizer(iwv.getContext(),
"holder", this);
// 此处须要将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获得的apiKey和secretKey
speechSynthesizer.setApiKey("your-apiKey", "your-secretKey");
speechSynthesizer.setAudioStreamType(AudioManager.STREAM_MUSIC);
//setVolumeControlStream(AudioManager.STREAM_MUSIC);
new Thread(new Runnable() {
@Override
public void run() {
setParams();
int ret = speechSynthesizer.speak(text);
if (ret != 0) {
Log.i(TAG,"开始合成器失败" );
}
}
}).start();
}
private void setParams() {
speechSynthesizer.setParam(SpeechSynthesizer.PARAM_SPEAKER, "0");
speechSynthesizer.setParam(SpeechSynthesizer.PARAM_VOLUME, "5");
speechSynthesizer.setParam(SpeechSynthesizer.PARAM_SPEED, "5");
speechSynthesizer.setParam(SpeechSynthesizer.PARAM_PITCH, "5");
speechSynthesizer.setParam(SpeechSynthesizer.PARAM_AUDIO_ENCODE, SpeechSynthesizer.AUDIO_ENCODE_AMR);
speechSynthesizer.setParam(SpeechSynthesizer.PARAM_AUDIO_RATE, SpeechSynthesizer.AUDIO_BITRATE_AMR_15K85);
}
@Override
public void onBufferProgressChanged(SpeechSynthesizer arg0, int arg1) {}
@Override
public void onCancel(SpeechSynthesizer arg0) {}
@Override
public void onError(SpeechSynthesizer arg0, SpeechError arg1) {}
@Override
public void onNewDataArrive(SpeechSynthesizer arg0, byte[] arg1, boolean arg2) {}
@Override
public void onSpeechFinish(SpeechSynthesizer arg0) {}
@Override
public void onSpeechPause(SpeechSynthesizer arg0) {}
@Override
public void onSpeechProgressChanged(SpeechSynthesizer arg0, int arg1) {}
@Override
public void onSpeechResume(SpeechSynthesizer arg0) {}
@Override
public void onSpeechStart(SpeechSynthesizer arg0) {}
@Override
public void onStartWorking(SpeechSynthesizer arg0) {}
@Override
public void onSynthesizeFinish(SpeechSynthesizer arg0) {}
}
前台javascript代码
document.addEventListener( "plusready", function()
{
var _BARCODE = 'kxdPlugins',
B = window.plus.bridge;
var kxdPlugins =
{
// 声明异步返回方法
speak : function (Argus, successCallback, errorCallback )
{
var success = typeof successCallback !== 'function' ? null : function(args)
{
successCallback(args);
},
fail = typeof errorCallback !== 'function' ? null : function(code)
{
errorCallback(code);
};
callbackID = B.callbackId(success, fail);
// 通知Native层plugintest扩展插件运行”speak”方法
return B.exec(_BARCODE, "speak", [callbackID, Argus]);
}
};
window.plus.kxdPlugins = kxdPlugins;
}, true );
使用方式
和普通的API调用方式同样:plus.kxdPlugins.speak("这是语音播放的文字内容");
注意事项
1.在/assets/data/properties.xml中添加插件权限。
2.在/assets/apps/appid/www/manifest.json 中添加对应插件权限。
3.在/AndroidManifest.xml 文件中添加对应的百度语音合成权限。
三,
前台javascript代码
Argus]); } }; window.plus.kxdPlugins = kxdPlugins; }, true );
和普通的API调用方式同样:plus.kxdPlugins.speak("这是语音播放的文字内容");
1.在/assets/data/properties.xml中添加插件权限。
2.在/assets/apps/appid/www/manifest.json 中添加对应插件权限。
3.在/AndroidManifest.xml 文件中添加对应的百度语音合成权限。
这是调用百度语音API
四,Android语音
code:
public class TextToSpeechDemo implements TextToSpeech.OnInitListener {
private final TextToSpeech mTextToSpeech;//TTS对象
private final ConcurrentLinkedQueue mBufferedMessages;//消息队列
private Context mContext;
private boolean mIsReady;//标识符
public TextToSpeechDemo(Context context){
this.mContext=context;//获取上下文
this.mBufferedMessages=new ConcurrentLinkedQueue();//实例化队列
this.mTextToSpeech=new TextToSpeech(this.mContext,this);//实例化TTS
}
//初始化TTS引擎
@Override
public void onInit(int status) {
Log.i("TextToSpeechDemo",String.valueOf(status));
if(status==TextToSpeech.SUCCESS){
int result = this.mTextToSpeech.setLanguage(Locale.CHINA);//设置识别语音为中文
synchronized (this){
this.mIsReady=true;//设置标识符为true
for(String bufferedMessage : this.mBufferedMessages){
speakText(bufferedMessage);//读语音
}
this.mBufferedMessages.clear();//读完后清空队列
}
}
}
//释放资源
public void release(){
synchronized (this){
this.mTextToSpeech.shutdown();
this.mIsReady=false;
}
}
//更新消息队列,或者读语音
public void notifyNewMessage(String lanaugh){
String message=lanaugh;
synchronized (this){
if(this.mIsReady){
speakText(message);
}else{
this.mBufferedMessages.add(message);
}
}
}
//读语音处理
private void speakText(String message){
Log.i("liyuanjinglyj",message);
HashMap params=new HashMap();
params.put(TextToSpeech.Engine.KEY_PARAM_STREAM,"STREAM_NOTIFICATION");//设置播放类型(音频流类型)
this.mTextToSpeech.speak(message, TextToSpeech.QUEUE_ADD, params);//将这个发音任务添加当前任务以后
this.mTextToSpeech.playSilence(100,TextToSpeech.QUEUE_ADD,params);//间隔多长时间
}
}
固然通常手机是不支持中文的能够百度下载讯飞TTS引擎后在测试。见最后附图。
因为TTS引擎的初始化是异步的,因此在执行实际的文字转语音以前须要把消息放到队列中。
能够给TTS引擎发送多个参数。前面展现了如何决定口语消息使用的音频流。在这种状况下,通知声音也使用相同的音频流。
最后,若是处理连续多个消息,最好在每一个消息结束后暂停一会在播放下一个消息。这样作会清楚的告诉用户消息的结束和开始。
效果图:
五,安装语音库,假如要中文发音,使用科大讯飞语音3.0
code:
package com.example.tts;
import java.util.Locale;
import android.speech.tts.TextToSpeech;
import android.support.v7.app.ActionBarActivity;
import android.os.Bundle;
import android.view.Menu;
import android.view.MenuItem;
import android.widget.Toast;
public class MainActivity extends ActionBarActivity
implements TextToSpeech.OnInitListener{
TextToSpeech textToSpeech = null; @Override
protected void onCreate(Bundle savedInstanceState) {
super.onCreate(savedInstanceState); setContentView(R.layout.activity_main); textToSpeech = new TextToSpeech(this, this);
textToSpeech.speak("此处无声", TextToSpeech.QUEUE_ADD, null); } @Override
public boolean onCreateOptionsMenu(Menu menu) { getMenuInflater().inflate(R.menu.main, menu);
return true; }
protected void onDestroy() { super.onDestroy();
if (textToSpeech!=null) { textToSpeech.shutdown(); } } @Override
public boolean onOptionsItemSelected(MenuItem item) {
int id = item.getItemId();
if (id == R.id.action_settings) {
textToSpeech.speak("此处有声", TextToSpeech.QUEUE_FLUSH, null);
return true; }
return super.onOptionsItemSelected(item); } @Override
public void onInit(int status) { // TODO Auto-generated method stub
if (status == TextToSpeech.SUCCESS) {
int result = textToSpeech.setLanguage(Locale.CHINESE);
if (result == TextToSpeech.LANG_MISSING_DATA || result == TextToSpeech.LANG_NOT_SUPPORTED ||
result == TextToSpeech.ERROR) {
Toast.makeText(this, "数据丢失或语言不支持",
Toast.LENGTH_SHORT).show(); }
if (result == TextToSpeech.LANG_AVAILABLE) {
Toast.makeText(this, "支持该语言", Toast.LENGTH_SHORT).show(); }
Toast.makeText(this, "初始化成功", Toast.LENGTH_SHORT).show(); } } }
无需任何权限,这里有个问题,就是在动态建立一个对象以后,在onCreate里面调用speak方法,并不能发出声音。
能够把文字保存为语音文件,也能够读取语音文件
public void saveToFile(TextToSpeech speech,String text,String file) { String destFileName = "/sdcard/tts/"+file+".wav"; speech.synthesizeToFile(text, null, destFileName); } public void readFromFile(TextToSpeech speech,String file) { String destFileName = "/sdcard/tts/"+file+".wav"; speech.addSpeech("2", destFileName); speech.speak("2", TextToSpeech.QUEUE_ADD, null); }
这样就能够了。
接下来说一下如何实现语音识别
语音识别首先能够考虑使用科大讯飞,可是目前使用必需要联网,并且申请一个APPID。
用起来不难
一、语音合成功能
private SpeechSynthesizer speechSynthesizer;
SpeechUser.getUser().login(MainActivity.this, null, null, "appid=54d304cf", null);
speechSynthesizer = SpeechSynthesizer.createSynthesizer(this);
speechSynthesizer.setParameter(SpeechConstant.VOICE_NAME, "xiaoyan");
speechSynthesizer.setParameter(SpeechConstant.SPEED, "50");
speechSynthesizer.setParameter(SpeechConstant.VOLUME, "50");
speechSynthesizer.setParameter(SpeechConstant.PITCH, "50");
String text = editText.getText().toString();speechSynthesizer.startSpeaking(text, null);
注册--设置--播放。文字转语音功能须要联网,但不须要正确的APPID,能够直接使用。可是语音识别就须要联网和正确的APPID
二、语音识别
private RecognizerDialog recognizerDialog;
SpeechUser.getUser().login(MainActivity.this, null, null, "appid=54d304cf", null);
recognizerDialog = new RecognizerDialog(this);
recognizerDialog.setParameter(SpeechConstant.DOMAIN, "iat");
recognizerDialog.setParameter(SpeechConstant.SAMPLE_RATE, "16000");
//显示Dialog
recognizerDialog.setListener(dialogListener);
recognizerDialog.show();
private RecognizerDialogListener dialogListener = new RecognizerDialogListener() {
//识别结果回调
@Override
public void onResult(RecognizerResult arg0, boolean arg1) {
// TODO Auto-generated method stub
String text = JsonParser.parseIatResult(arg0.getResultString());
editText.append(text);
editText.setSelection(editText.length());
}
//识别结束回调
@Override
public void onError(SpeechError arg0) {
// TODO Auto-generated method stub
}
};
package com.example.viocedemo;
import org.json.JSONArray;
import org.json.JSONObject;
import org.json.JSONTokener;
import android.text.TextUtils; /** * 对云端返回的Json结果进行解析 * @author iFlytek * @since 20131211 */ public class JsonParser { /** * 听写结果的Json格式解析 * @param json * @return */
public static String parseIatResult(String json) {
if(TextUtils.isEmpty(json))
return ""; StringBuffer ret = new StringBuffer();
try { JSONTokener tokener = new JSONTokener(json);
JSONObject joResult = new JSONObject(tokener);
JSONArray words = joResult.getJSONArray("ws");
for (int i = 0; i < words.length(); i++) {
JSONArray items = words.getJSONObject(i).getJSONArray("cw");
JSONObject obj = items.getJSONObject(0);
ret.append(obj.getString("w")); } }
catch (Exception e) { e.printStackTrace(); }
return ret.toString(); } }
这样就能够实现语音输入了,很是简单。
PHP文字转语音,这类的文章少,可能要集成PHP第三方插件来实现。
Android默认没有安装TTS数据包,没法文字转语音,而在设置里推荐的语音包是Pico TTS,并不支持中文,若是须要读中文,须要下载另外的第三方语音包,如:eSpeak,Svox,我的建议Svox,eSpeak很是生硬,并且不少汉字都读不出,不支持中英文混读(英文当拼音读)。下载连接本身Google下吧,不少软件市场都有下载。下载安装后打开设置-语音输入和输出-文字转语音设置,勾选Svox Classic TTS,语言选择中文或广东话(Svox安装后还要下载安装相应的语言包的,相似插件)。
中文普通话语音包:http://www.nduoa.com/web/soft/show/id/4954
粤语语音包:http://www.nduoa.com/web/soft/show/id/4955
实现普通话,粤语
文字转语音编程能够实现更多地方语言的扩展,好比四川话,闽南语,江西话等,办法是有的,能够提供更多的语音扩展。。。。。。
文字转语音,相信再也不是英文,汉语,韩语,也许是家乡话更亲切。
拭目以待
end