一、开通阿里云TTS服务
登录阿里云，选择菜单：产品->人工智能->语音合成
点击“申请开通”，然后在“管理控制台”创建一个项目
复制 appkey
注意，token只有1天有效，所以需要通过接口去定时获取
二、对接语音合成api接口
由于sdk需要引入很多第三方jar包，所以建议对接RESTful API
copy接口文档里的demo代码，把申请到token和appkey粘贴进去，可以直接运行，demo会生成一个syAudio.wav文件，使用语言播放器直接播放就可以。
根据文档提示需要在工程中引入三个jar包：
<dependency>
<groupId>com.squareup.okhttp3</groupId>
<artifactId>okhttp</artifactId>
<version>3.9.1</version>
</dependency>
<!-- http://mvnrepository.com/artifact/com.alibaba/fastjson -->
<dependency>
<groupId>com.alibaba</groupId>
<artifactId>fastjson</artifactId>
<version>1.2.42</version>
</dependency>
<!-- 获取token使用 -->
<dependency>
<groupId>com.aliyun</groupId>
<artifactId>aliyun-java-sdk-core</artifactId>
<version>3.7.1</version>
</dependency>
语音生成工具类：
package com.hsoft.web.util;
import java.io.UnsupportedEncodingException;
import java.net.URLEncoder;
import org.apache.commons.lang3.StringUtils;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import com.alibaba.fastjson.JSONObject;
import com.hsoft.commutil.props.PropertiesUtil;
import okhttp3.MediaType;
import okhttp3.OkHttpClient;
import okhttp3.Request;
import okhttp3.RequestBody;
import okhttp3.Response;
public class SpeechRestfulUtil {
private static Logger logger = LoggerFactory.getLogger(SpeechRestfulUtil.class);
private String accessToken;
private String appkey;
private static SpeechRestfulUtil getInstance() {
String appkey = PropertiesUtil.getProperty("aliyun.voice.appkey");
String token = AliTokenUtil.getToken();
return new SpeechRestfulUtil(appkey, token);
}
private SpeechRestfulUtil(String appkey, String token) {
this.appkey = appkey;
this.accessToken = token;
}
/**
* HTTPS GET 请求
*/
private byte[] processGETRequet(String text, String format, int sampleRate) {
/**
* 设置HTTPS GET请求
* 1.使用HTTPS协议
* 2.语音识别服务域名：nls-gateway.cn-shanghai.aliyuncs.com
* 3.语音识别接口请求路径：/stream/v1/tts
* 4.设置必须请求参数：appkey、token、text、format、sample_rate
* 5.设置可选请求参数：voice、volume、speech_rate、pitch_rate
*/
String url = "https://nls-gateway.cn-shanghai.aliyuncs.com/stream/v1/tts";
url = url + "?appkey=" + appkey;
url = url + "&token=" + accessToken;
url = url + "&text=" + text;
url = url + "&format=" + format;
url = url + "&sample_rate=" + String.valueOf(sampleRate);
// voice 发音人，可选，默认是xiaoyun
// url = url + "&voice=" + "xiaoyun";
// volume 音量，范围是0~100，可选，默认50
// url = url + "&volume=" + String.valueOf(50);
// speech_rate 语速，范围是-500~500，可选，默认是0
url = url + "&speech_rate=" + String.valueOf(100);
// pitch_rate 语调，范围是-500~500，可选，默认是0
// url = url + "&pitch_rate=" + String.valueOf(0);
// System.out.println("URL: " + url);
/**
* 发送HTTPS GET请求，处理服务端的响应
*/
Request request = new Request.Builder()
.url(url)
.get()
.build();
byte[] bytes=null;
try {
OkHttpClient client = new OkHttpClient();
Response response = client.newCall(request).execute();
String contentType = response.header("Content-Type");
if ("audio/mpeg".equals(contentType)) {
bytes =response.body().bytes();
// File f = new File(audioSaveFile);
// FileOutputStream fout = new FileOutputStream(f);
// fout.write(response.body().bytes());
// fout.close();
// System.out.println(f.getAbsolutePath());
logger.info("The GET SpeechRestful succeed!");
}
else {
// ContentType 为 null 或者为 "application/json"
String errorMessage = response.body().string();
logger.info("The GET SpeechRestful failed: " + errorMessage);
}
response.close();
} catch (Exception e) {
logger.error("processGETRequet",e);
}
return bytes;
}
/**
* HTTPS POST 请求
*/
private byte[] processPOSTRequest(String text, String audioSaveFile, String format, int sampleRate) {
/**
* 设置HTTPS POST请求
* 1.使用HTTPS协议
* 2.语音合成服务域名：nls-gateway.cn-shanghai.aliyuncs.com
* 3.语音合成接口请求路径：/stream/v1/tts
* 4.设置必须请求参数：appkey、token、text、format、sample_rate
* 5.设置可选请求参数：voice、volume、speech_rate、pitch_rate
*/
String url = "https://nls-gateway.cn-shanghai.aliyuncs.com/stream/v1/tts";
JSONObject taskObject = new JSONObject();
taskObject.put("appkey", appkey);
taskObject.put("token", accessToken);
taskObject.put("text", text);
taskObject.put("format", format);
taskObject.put("sample_rate", sampleRate);
// voice 发音人，可选，默认是xiaoyun
// taskObject.put("voice", "xiaoyun");
// volume 音量，范围是0~100，可选，默认50
// taskObject.put("volume", 50);
// speech_rate 语速，范围是-500~500，可选，默认是0
// taskObject.put("speech_rate", 0);
// pitch_rate 语调，范围是-500~500，可选，默认是0
// taskObject.put("pitch_rate", 0);
String bodyContent = taskObject.toJSONString();
// System.out.println("POST Body Content: " + bodyContent);
RequestBody reqBody = RequestBody.create(MediaType.parse("application/json"), bodyContent);
Request request = new Request.Builder()
.url(url)
.header("Content-Type", "application/json")
.post(reqBody)
.build();
byte[] bytes=null;
try {
OkHttpClient client = new OkHttpClient();
Response response = client.newCall(request).execute();
String contentType = response.header("Content-Type");
if ("audio/mpeg".equals(contentType)) {
bytes = response.body().bytes();
logger.info("The POST SpeechRestful succeed!");
}
else {
// ContentType 为 null 或者为 "application/json"
String errorMessage = response.body().string();
logger.info("The POST SpeechRestful failed: " + errorMessage);
}
response.close();
} catch (Exception e) {
logger.error("processPOSTRequest",e);
}
return bytes;
}
public static byte[] text2voice(String text) {
if (StringUtils.isBlank(text)) {
return null;
}
SpeechRestfulUtil demo = SpeechRestfulUtil.getInstance();
// String text = "会员收款87.12元";
// 采用RFC 3986规范进行urlencode编码
String textUrlEncode = text;
try {
textUrlEncode = URLEncoder.encode(textUrlEncode, "UTF-8")
.replace("+", "%20")
.replace("*", "%2A")
.replace("%7E", "~");
} catch (UnsupportedEncodingException e) {
logger.error("encode",e);
}
// String audioSaveFile = "syAudio.wav";
String format = "wav";
int sampleRate = 16000;
return demo.processGETRequet(textUrlEncode, format, sampleRate);
}
}
获取Token工具类
import org.apache.commons.lang3.StringUtils;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import com.alibaba.fastjson.JSON;
import com.alibaba.fastjson.JSONObject;
import com.aliyuncs.CommonRequest;
import com.aliyuncs.CommonResponse;
import com.aliyuncs.DefaultAcsClient;
import com.aliyuncs.IAcsClient;
import com.aliyuncs.http.MethodType;
import com.aliyuncs.http.ProtocolType;
import com.aliyuncs.profile.DefaultProfile;
import com.hsoft.commutil.props.PropertiesUtil;
public class AliTokenUtil {
private static Logger logger = LoggerFactory.getLogger(AliTokenUtil.class);
// 您的地域ID
private static final String REGIONID = "cn-shanghai";
// 获取Token服务域名
private static final String DOMAIN = "nls-meta.cn-shanghai.aliyuncs.com";
// API 版本
private static final String API_VERSION = "2019-02-28";
// API名称
private static final String REQUEST_ACTION = "CreateToken";
// 响应参数
private static final String KEY_TOKEN = "Token";
private static final String KEY_ID = "Id";
private static final String KEY_EXPIRETIME = "ExpireTime";
private static volatile String TOKEN = "";
private static volatile long EXPIRETIME = 0L;
public static String getToken() {
if (StringUtils.isNotBlank(TOKEN)) {
if (EXPIRETIME - System.currentTimeMillis() / 1000 > 3600) {
return TOKEN;
}
}
try {
String accessKeyId = PropertiesUtil.getProperty("aliyun.accessId");;
String accessKeySecret = PropertiesUtil.getProperty("aliyun.accessKey");;
// 创建DefaultAcsClient实例并初始化
DefaultProfile profile = DefaultProfile.getProfile(REGIONID, accessKeyId, accessKeySecret);
IAcsClient client = new DefaultAcsClient(profile);
CommonRequest request = new CommonRequest();
request.setDomain(DOMAIN);
request.setVersion(API_VERSION);
request.setAction(REQUEST_ACTION);
request.setMethod(MethodType.POST);
request.setProtocol(ProtocolType.HTTPS);
CommonResponse response = client.getCommonResponse(request);
logger.info(response.getData());
if (response.getHttpStatus() == 200) {
JSONObject result = JSON.parseObject(response.getData());
TOKEN = result.getJSONObject(KEY_TOKEN).getString(KEY_ID);
EXPIRETIME = result.getJSONObject(KEY_TOKEN).getLongValue(KEY_EXPIRETIME);
logger.info("获取到的Token： " + TOKEN + "，有效期时间戳(单位：秒): " + EXPIRETIME);
} else {
logger.info("获取Token失败！");
}
} catch (Exception e) {
logger.error("getToken error!", e);
}
return TOKEN;
}
}
三、集成websocket
当然，我们的目的不是得到一个音频文件，而是在web站点上可以直接听见声音。
为此，需要引入Websocket，将得到的音频资源直接推送到web页面上，然后使用FileReader对象直接播放
1、引入jar包
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-websocket</artifactId>
<exclusions>
<exclusion>
<groupId>org.slf4j</groupId>
<artifactId>log4j-over-slf4j</artifactId>
</exclusion>
<exclusion>
<groupId>org.hibernate</groupId>
<artifactId>hibernate-validator</artifactId>
</exclusion>
</exclusions>
</dependency>
2、创建Websocket处理类
public class VoiceHandler extends AbstractWebSocketHandler {
private static final Logger logger = LoggerFactory.getLogger(VoiceHandler.class);
@Override
public void afterConnectionEstablished(WebSocketSession session) throws Exception {
VoicePool.add(session);
}
@Override
public void afterConnectionClosed(WebSocketSession session, CloseStatus status) throws Exception {
VoicePool.remove(session);
}
@Override
protected void handleTextMessage(WebSocketSession session, TextMessage message) throws Exception {
logger.debug("receive Msg :" + message.getPayload());
TextMessage msg=new TextMessage(message.getPayload());
session.sendMessage(msg);
}
}
3、创建websocket连接池管理类
public class VoicePool {
private static final Logger logger = LoggerFactory.getLogger(VoicePool.class);
private static Map<String, WebSocketSession> pool = new ConcurrentHashMap<String, WebSocketSession>();
private static Map<Long, List<String>> userMap = new ConcurrentHashMap<Long, List<String>>();
private static final ExecutorService threadPool = Executors.newFixedThreadPool(50);
public static void add(WebSocketSession inbound) {
pool.put(inbound.getId(), inbound);
Map<String, String> map = ParamUtil.parser(inbound.getUri().getQuery());
Long companyId = Long.valueOf(map.get("companyId"));
logger.info("add companyId:{}", companyId);
List<String> lstInBound = null;
if (companyId != null) {
lstInBound = userMap.get(companyId);
if (lstInBound == null) {
lstInBound = new ArrayList<String>();
userMap.put(companyId, lstInBound);
}
lstInBound.add(inbound.getId());
}
logger.info("add connetion {},total size {}", inbound.getId(), pool.size());
}
public static void remove(WebSocketSession socket) {
String sessionId = socket.getId();
List<String> lstInBound = null;
Map<String, String> map = ParamUtil.parser(socket.getUri().getQuery());
Long companyId = Long.valueOf(map.get("companyId"));
logger.info("remove companyId:{}", companyId);
if (StringUtils.isNotBlank(sessionId)) {
if (companyId != null) {
lstInBound = userMap.get(companyId);
if (lstInBound != null) {
lstInBound.remove(sessionId);
if (lstInBound.isEmpty()) {
userMap.remove(companyId);
}
}
}
}
pool.remove(sessionId);
logger.info("remove connetion {},total size {}", sessionId, pool.size());
}
/** 推送信息 */
public static void broadcast(VoiceMsgVo vo) {
Long companyId = vo.getCompanyId();
if (companyId == null || companyId == 0L) {
return;
}
List<String> lstInBoundId = userMap.get(companyId);
if (lstInBoundId == null || lstInBoundId.isEmpty()) {
return;
}
byte[] bytes = SpeechRestfulUtil.text2voice(vo.getText());
if (bytes == null) {
return;
}
threadPool.execute(() -> {
try {
for (String id : lstInBoundId) {
// 发送给指定用户
WebSocketSession connection = pool.get(id);
if (connection != null) {
synchronized (connection) {
BinaryMessage msg = new BinaryMessage(bytes);
connection.sendMessage(msg);
}
}
}
} catch (Exception e) {
logger.error("broadcast error: companyId:{}", companyId, e);
}
});
}
}
消息对象bean
public class VoiceMsgVo {
private String text;
private Long companyId;
}
4、Websocket配置
@Configuration
@EnableWebSocket
public class WebSocketConfig implements WebSocketConfigurer {
@Override
public void registerWebSocketHandlers(WebSocketHandlerRegistry registry) {
registry.addHandler(voiceHandler(), "/ws/voice").setAllowedOrigins("*");
}
@Bean
public VoiceHandler voiceHandler() {
return new VoiceHandler();
}
}
5、前端js处理
随便创建也页面，引入下面的js
var audioContext = new (window.AudioContext || window.webkitAudioContext)();
var Chat = {};
Chat.socket = null;
Chat.connect = (function(host) {
if ("WebSocket" in window) {
Chat.socket = new WebSocket(host);
} else if ("MozWebSocket" in window) {
Chat.socket = new MozWebSocket(host);
} else {
Console.log("Error: WebSocket is not supported by this browser.");
return;
}
Chat.socket.onopen = function() {
Console.log("Info: 语音播报已启动.");
// 心跳检测重置
heartCheck.reset().start(Chat.socket);
};
Chat.socket.onclose = function() {
Console.log("Info: 语音播报已关闭.");
};
Chat.socket.onmessage = function(message) {
heartCheck.reset().start(Chat.socket);
if (message.data == null || message.data == '' || "HeartBeat" == message.data){
//心跳消息
return;
}
var reader = new FileReader();
reader.onload = function(evt) {
if (evt.target.readyState == FileReader.DONE) {
audioContext.decodeAudioData(evt.target.result,
function(buffer) {
// 解码成pcm流
var audioBufferSouceNode = audioContext
.createBufferSource();
audioBufferSouceNode.buffer = buffer;
audioBufferSouceNode
.connect(audioContext.destination);
audioBufferSouceNode.start(0);
}, function(e) {
console.log(e);
});
}
};
reader.readAsArrayBuffer(message.data);
};
});
Chat.initialize = function() {
Chat.companyId = _currCompanyId;
if (window.location.protocol == "http:") {
Chat.connect("ws://" + window.location.host + "/ws/voice?companyId="+Chat.companyId);
} else {
Chat.connect("wss://" + window.location.host + "/ws/voice?companyId="+Chat.companyId);
}
};
Chat.sendMessage = (function() {
var message = document.getElementById("chat").value;
if (message != "") {
Chat.socket.send(message);
document.getElementById("chat").value = "";
}
});
var Console = {};
Console.log = (function(message) {
var _console=document.getElementById("console");
if (_console==null || _console==undefined){
console.log(message);
return;
}
var p=document.createElement("p");
p.style.wordWrap="break-word";
p.innerHTML=message;
_console.appendChild(p);
while(_console.childNodes.length>25)
{
_console.removeChild(_console.firstChild);
}
_console.scrollTop=_console.scrollHeight;
});
Chat.initialize();
//心跳检测
var heartCheck = {
timeout : 60000,// 60秒
timeoutObj : null,
serverTimeoutObj : null,
reset : function() {
clearTimeout(this.timeoutObj);
clearTimeout(this.serverTimeoutObj);
return this;
},
start : function(ws) {
var self = this;
this.timeoutObj = setTimeout(function() {
// 这里发送一个心跳，后端收到后，返回一个心跳消息，
// onmessage拿到返回的心跳就说明连接正常
// console.log('start heartCheck');
ws.send("HeartBeat");
self.serverTimeoutObj = setTimeout(function() {// 如果超过一定时间还没重置，说明后端主动断开了
ws.close();// 如果onclose会执行reconnect，我们执行ws.close()就行了.如果直接执行reconnect
// 会触发onclose导致重连两次
}, self.timeout)
}, this.timeout)
}
}
四、启动工程测试
启动工程，从后台发送一段消息
VoiceMsgVo vo = new VoiceMsgVo();
vo.setCompanyId(1L);
vo.setText("今天天气真好！我出去散步了");
VoicePool.broadcast(vo);