├── .gitignore
├── pom.xml
├── readme.md
├── src
    ├── main
    │   ├── java
    │   │   └── com
    │   │   │   └── taoing
    │   │   │       └── ttsserver
    │   │   │           ├── TtsServerApplication.java
    │   │   │           ├── controller
    │   │   │               └── TTSController.java
    │   │   │           ├── service
    │   │   │               └── TTSService.java
    │   │   │           ├── utils
    │   │   │               └── Tool.java
    │   │   │           ├── vo
    │   │   │               ├── GeneralResponse.java
    │   │   │               └── TextTTSReq.java
    │   │   │           └── ws
    │   │   │               ├── TTSConfig.java
    │   │   │               ├── TTSWebSocket.java
    │   │   │               ├── TTSWebSocketListener.java
    │   │   │               └── WebSocketManager.java
    │   └── resources
    │   │   ├── application-dev.yml
    │   │   ├── application-pro.yml
    │   │   ├── application-test.yml
    │   │   ├── application.yml
    │   │   └── logback-spring.xml
    └── test
    │   └── java
    │       └── com
    │           └── taoing
    │               └── ttsserver
    │                   └── TtsServerApplicationTests.java
└── startup.sh


/.gitignore:
--------------------------------------------------------------------------------
 1 | target/
 2 | 
 3 | ### IntelliJ IDEA ###
 4 | .idea
 5 | *.iws
 6 | *.iml
 7 | *.ipr
 8 | 
 9 | logs
10 | 


--------------------------------------------------------------------------------
/pom.xml:
--------------------------------------------------------------------------------
 1 | <?xml version="1.0" encoding="UTF-8"?>
 2 | <project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
 3 |          xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 https://maven.apache.org/xsd/maven-4.0.0.xsd">
 4 |     <modelVersion>4.0.0</modelVersion>
 5 |     <parent>
 6 |         <groupId>org.springframework.boot</groupId>
 7 |         <artifactId>spring-boot-starter-parent</artifactId>
 8 |         <version>2.5.4</version>
 9 |         <relativePath/> <!-- lookup parent from repository -->
10 |     </parent>
11 |     <groupId>com.taoing</groupId>
12 |     <artifactId>tts-server</artifactId>
13 |     <version>1.0.0</version>
14 |     <name>tts-server</name>
15 |     <description>microsoft edge tts(xiaoxiao)</description>
16 |     <properties>
17 |         <java.version>1.8</java.version>
18 |     </properties>
19 |     <dependencies>
20 |         <dependency>
21 |             <groupId>org.springframework.boot</groupId>
22 |             <artifactId>spring-boot-starter-web</artifactId>
23 |         </dependency>
24 | 
25 |         <dependency>
26 |             <groupId>com.squareup.okhttp3</groupId>
27 |             <artifactId>okhttp</artifactId>
28 |             <version>4.9.1</version>
29 |         </dependency>
30 | 
31 |         <dependency>
32 |             <groupId>org.projectlombok</groupId>
33 |             <artifactId>lombok</artifactId>
34 |             <optional>true</optional>
35 |         </dependency>
36 | 
37 |         <dependency>
38 |             <groupId>org.springframework.boot</groupId>
39 |             <artifactId>spring-boot-configuration-processor</artifactId>
40 |             <optional>true</optional>
41 |         </dependency>
42 | 
43 |         <dependency>
44 |             <groupId>org.springframework.boot</groupId>
45 |             <artifactId>spring-boot-starter-test</artifactId>
46 |             <scope>test</scope>
47 |         </dependency>
48 |     </dependencies>
49 | 
50 |     <build>
51 |         <plugins>
52 |             <plugin>
53 |                 <groupId>org.springframework.boot</groupId>
54 |                 <artifactId>spring-boot-maven-plugin</artifactId>
55 |             </plugin>
56 |         </plugins>
57 |     </build>
58 | 
59 | </project>
60 | 


--------------------------------------------------------------------------------
/readme.md:
--------------------------------------------------------------------------------
1 | # TTS Server
2 | 
3 | 利用微软edge的朗读功能, 文本合成语音. 可配合爱阅书香app使用, 实现"微软晓晓"语音在线听书...
4 | 
5 | 了解更多微软tts:
6 | https://azure.microsoft.com/zh-cn/services/cognitive-services/text-to-speech
7 | 


--------------------------------------------------------------------------------
/src/main/java/com/taoing/ttsserver/TtsServerApplication.java:
--------------------------------------------------------------------------------
 1 | package com.taoing.ttsserver;
 2 | 
 3 | import org.springframework.boot.SpringApplication;
 4 | import org.springframework.boot.autoconfigure.SpringBootApplication;
 5 | 
 6 | @SpringBootApplication
 7 | public class TtsServerApplication {
 8 | 
 9 |     public static void main(String[] args) {
10 |         SpringApplication.run(TtsServerApplication.class, args);
11 |     }
12 | 
13 | }
14 | 


--------------------------------------------------------------------------------
/src/main/java/com/taoing/ttsserver/controller/TTSController.java:
--------------------------------------------------------------------------------
 1 | package com.taoing.ttsserver.controller;
 2 | 
 3 | import com.taoing.ttsserver.service.TTSService;
 4 | import com.taoing.ttsserver.vo.GeneralResponse;
 5 | import com.taoing.ttsserver.vo.TextTTSReq;
 6 | import lombok.extern.slf4j.Slf4j;
 7 | import org.springframework.beans.factory.annotation.Autowired;
 8 | import org.springframework.util.StringUtils;
 9 | import org.springframework.web.bind.annotation.GetMapping;
10 | import org.springframework.web.bind.annotation.PostMapping;
11 | import org.springframework.web.bind.annotation.RequestMapping;
12 | import org.springframework.web.bind.annotation.RestController;
13 | 
14 | import javax.servlet.http.HttpServletRequest;
15 | import javax.servlet.http.HttpServletResponse;
16 | 
17 | @Slf4j
18 | @RequestMapping(value = "/tests")
19 | @RestController
20 | public class TTSController {
21 | 
22 |     @Autowired
23 |     private TTSService service;
24 | 
25 |     /**
26 |      * 合成文本语音
27 |      * 接口不支持多线程, 因此, 一个key只支持一个用户调用
28 |      * 可用key: "xxttsa01", "xxttsa02", "xxttsa03"
29 |      * @param response
30 |      * @param param
31 |      */
32 |     @PostMapping(value = "/text2Speech")
33 |     public void text2Speech(HttpServletResponse response, TextTTSReq param) {
34 |         if (log.isDebugEnabled()) {
35 |             log.debug("req: {}", param.toString());
36 |         }
37 |         if (!StringUtils.hasLength(param.getKey())) {
38 |             response.setStatus(HttpServletResponse.SC_BAD_REQUEST);
39 |             return;
40 |         }
41 |         if (!StringUtils.hasLength(param.getText())) {
42 |             return;
43 |         }
44 |         this.service.synthesisAudio(response, param);
45 |     }
46 | 
47 |     @GetMapping(value = "/test")
48 |     public GeneralResponse<Void> apiTest(HttpServletRequest request) {
49 |         log.info("request uri: {}", request.getRequestURI() + "?" + request.getQueryString());
50 |         return new GeneralResponse<>();
51 |     }
52 | }
53 | 


--------------------------------------------------------------------------------
/src/main/java/com/taoing/ttsserver/service/TTSService.java:
--------------------------------------------------------------------------------
 1 | package com.taoing.ttsserver.service;
 2 | 
 3 | import com.taoing.ttsserver.vo.TextTTSReq;
 4 | import com.taoing.ttsserver.ws.TTSWebSocket;
 5 | import com.taoing.ttsserver.ws.WebSocketManager;
 6 | import lombok.extern.slf4j.Slf4j;
 7 | import org.springframework.beans.factory.annotation.Autowired;
 8 | import org.springframework.stereotype.Service;
 9 | 
10 | import javax.servlet.http.HttpServletResponse;
11 | 
12 | @Slf4j
13 | @Service
14 | public class TTSService {
15 | 
16 |     @Autowired
17 |     private WebSocketManager webSocketManager;
18 | 
19 |     /**
20 |      * 合成文本语音
21 |      * @param response
22 |      * @param param
23 |      */
24 |     public void synthesisAudio(HttpServletResponse response, TextTTSReq param) {
25 |         TTSWebSocket client = webSocketManager.getClient(param.getKey());
26 |         if (client == null) {
27 |             log.warn("请求key非法");
28 |             response.setStatus(HttpServletResponse.SC_BAD_REQUEST);
29 |             return;
30 |         }
31 |         client.send(param.getText());
32 |         client.writeAudioStream(response);
33 |     }
34 | }
35 | 


--------------------------------------------------------------------------------
/src/main/java/com/taoing/ttsserver/utils/Tool.java:
--------------------------------------------------------------------------------
 1 | package com.taoing.ttsserver.utils;
 2 | 
 3 | public class Tool {
 4 | 
 5 |     /**
 6 |      * 移除字符串首尾空字符的高效方法(利用ASCII值判断,包括全角空格)
 7 |      * @param s
 8 |      * @return
 9 |      */
10 |     public static String fixTrim(String s) {
11 |         if (s == null || s.isEmpty()) {
12 |             return "";
13 |         }
14 |         int start = 0;
15 |         int len = s.length();
16 |         int end = len - 1;
17 |         while (start < end && (s.charAt(start) <= 0x20 || s.charAt(start) == '　')) {
18 |             ++start;
19 |         }
20 |         while (start < end && (s.charAt(end) <= 0x20 || s.charAt(end) == '　')) {
21 |             --end;
22 |         }
23 |         ++end;
24 |         return (start > 0 || end < len) ? s.substring(start, end) : s;
25 | 
26 |     }
27 | }
28 | 


--------------------------------------------------------------------------------
/src/main/java/com/taoing/ttsserver/vo/GeneralResponse.java:
--------------------------------------------------------------------------------
 1 | package com.taoing.ttsserver.vo;
 2 | 
 3 | import lombok.Data;
 4 | 
 5 | @Data
 6 | public class GeneralResponse<T> {
 7 |     private boolean success;
 8 | 
 9 |     private String code;
10 | 
11 |     private String message;
12 | 
13 |     private String errorMessage;
14 | 
15 |     private T data;
16 | 
17 | 
18 |     public GeneralResponse() {
19 |         this.success = true;
20 |         this.code = "1";
21 |         this.message = "请求成功";
22 |     }
23 | 
24 |     public GeneralResponse<T> error(String code, String errorMessage) {
25 |         this.success = false;
26 |         this.code = code;
27 |         this.message = "请求失败";
28 |         this.errorMessage = errorMessage;
29 |         return this;
30 |     }
31 | 
32 | }
33 | 


--------------------------------------------------------------------------------
/src/main/java/com/taoing/ttsserver/vo/TextTTSReq.java:
--------------------------------------------------------------------------------
 1 | package com.taoing.ttsserver.vo;
 2 | 
 3 | import lombok.Data;
 4 | 
 5 | @Data
 6 | public class TextTTSReq {
 7 | 
 8 |     /**
 9 |      * 文本
10 |      */
11 |     private String text;
12 | 
13 |     /**
14 |      * tts服务key
15 |      */
16 |     private String key;
17 | }
18 | 


--------------------------------------------------------------------------------
/src/main/java/com/taoing/ttsserver/ws/TTSConfig.java:
--------------------------------------------------------------------------------
 1 | package com.taoing.ttsserver.ws;
 2 | 
 3 | import lombok.Getter;
 4 | import lombok.Setter;
 5 | import org.springframework.boot.context.properties.ConfigurationProperties;
 6 | import org.springframework.context.annotation.Configuration;
 7 | 
 8 | import java.text.SimpleDateFormat;
 9 | import java.util.Date;
10 | import java.util.Locale;
11 | 
12 | @Getter
13 | @Setter
14 | @ConfigurationProperties(prefix = "tts.edge")
15 | @Configuration
16 | public class TTSConfig {
17 | 
18 |     /**
19 |      * 微软可用声音tts列表查询url
20 |      */
21 |     private String voiceListUrl;
22 | 
23 |     /**
24 |      * 语音合成websocket服务
25 |      */
26 |     private String wssUrl;
27 | 
28 |     /**
29 |      * token
30 |      */
31 |     private String trustedClientToken;
32 | 
33 |     /**
34 |      * 默认使用的声音tts(微软晓晓)
35 |      */
36 |     private String voiceName;
37 | 
38 |     /**
39 |      * 默认使用的声音tts(微软晓晓简写)
40 |      */
41 |     private String voiceShortName;
42 | 
43 |     /**
44 |      * 声音编码
45 |      */
46 |     private String codec;
47 | 
48 |     /**
49 |      * 区域
50 |      */
51 |     private String local;
52 | 
53 |     private boolean sentenceBoundaryEnabled;
54 | 
55 |     private boolean wordBoundaryEnabled;
56 | 
57 |     /**
58 |      * 语音合成配置
59 |      * @return
60 |      */
61 |     public String getSynthesizeConfig() {
62 |         String prefix = "X-Timestamp:+" + getTime() + "\r\n" +
63 |                 "Content-Type:application/json; charset=utf-8\r\n" +
64 |                 "Path:speech.config\r\n\r\n";
65 |         String config = "{\"context\":{\"synthesis\":{\"audio\":{\"metadataoptions\":{\"sentenceBoundaryEnabled\":\"%s\",\"wordBoundaryEnabled\":\"%s\"},\"outputFormat\":\"%s\"}}}}\r\n";
66 |         config = String.format(config, sentenceBoundaryEnabled ? "true" : "false",
67 |                 wordBoundaryEnabled ? "true" : "false", codec);
68 |         return prefix + config;
69 |     }
70 | 
71 |     public String getTime() {
72 |         SimpleDateFormat sdf = new SimpleDateFormat("EEE MMM dd yyyy HH:mm:ss 'GMT'Z (中国标准时间)", Locale.ENGLISH);
73 |         Date date = new Date();
74 |         return sdf.format(date);
75 |     }
76 | }
77 | 


--------------------------------------------------------------------------------
/src/main/java/com/taoing/ttsserver/ws/TTSWebSocket.java:
--------------------------------------------------------------------------------
  1 | package com.taoing.ttsserver.ws;
  2 | 
  3 | import com.taoing.ttsserver.utils.Tool;
  4 | import lombok.Getter;
  5 | import lombok.Setter;
  6 | import lombok.extern.slf4j.Slf4j;
  7 | import okhttp3.OkHttpClient;
  8 | import okhttp3.Request;
  9 | import okhttp3.WebSocket;
 10 | import okio.Buffer;
 11 | import okio.ByteString;
 12 | 
 13 | import javax.servlet.http.HttpServletResponse;
 14 | import java.io.IOException;
 15 | import java.io.OutputStream;
 16 | import java.util.UUID;
 17 | 
 18 | /**
 19 |  * websocket client包装
 20 |  */
 21 | @Slf4j
 22 | @Getter
 23 | @Setter
 24 | public class TTSWebSocket {
 25 | 
 26 |     private static int serialNum = 1;
 27 | 
 28 |     private Integer id;
 29 | 
 30 |     /**
 31 |      * 编码(自定义取值, 用以识别websocket client)
 32 |      */
 33 |     private String code;
 34 | 
 35 |     /**
 36 |      * websocket client
 37 |      */
 38 |     private WebSocket webSocket;
 39 | 
 40 |     /**
 41 |      * websocket是否可用
 42 |      */
 43 |     private boolean available;
 44 | 
 45 |     /**
 46 |      * tts配置
 47 |      */
 48 |     private TTSConfig ttsConfig;
 49 | 
 50 |     /**
 51 |      * 是否正在合成语音
 52 |      */
 53 |     private boolean synthesizing;
 54 | 
 55 |     /**
 56 |      * 音频的mime类型
 57 |      */
 58 |     private String mime;
 59 | 
 60 |     /**
 61 |      * 音频数据缓存
 62 |      */
 63 |     private Buffer buffer;
 64 | 
 65 |     public TTSWebSocket() {
 66 |     }
 67 | 
 68 |     public TTSWebSocket(String code, TTSConfig ttsConfig) {
 69 |         this.id = this.getSerialNum();
 70 |         this.code = code;
 71 |         this.available = true;
 72 |         this.ttsConfig = ttsConfig;
 73 |         this.webSocket = this.createWs();
 74 |     }
 75 | 
 76 |     @Override
 77 |     public String toString() {
 78 |         return "TTSWebSocket{" +
 79 |                 "id=" + id +
 80 |                 ", code='" + code + '\'' +
 81 |                 ", available=" + available +
 82 |                 ", synthesizing=" + synthesizing +
 83 |                 ", mime='" + mime + '\'' +
 84 |                 '}';
 85 |     }
 86 | 
 87 |     /**
 88 |      * 部分属性初始化
 89 |      */
 90 |     public void init() {
 91 |         this.mime = null;
 92 |         this.buffer = new Buffer();
 93 |     }
 94 | 
 95 |     public void send(String originText) {
 96 |         this.init();
 97 |         String text = Tool.fixTrim(originText);
 98 |         // 替换一些会导致不返回数据的特殊字符\p{N}
 99 |         /**
100 |          * \s matches any whitespace character (equivalent to [\r\n\t\f\v ])
101 |          * \p{P} matches any kind of punctuation character
102 |          * \p{Z} matches any kind of whitespace or invisible separator
103 |          * \p{S} matches any math symbols, currency signs, dingbats, box-drawing characters, etc
104 |          */
105 |         String temp = text.replaceAll("[\\s\\p{P}\\p{Z}\\p{S}]", "");
106 |         if (temp.length() < 1) {
107 |             // 可识别长度为0
108 |             this.synthesizing = false;
109 |             return;
110 |         }
111 |         text = this.escapeXmlChar(text);
112 |         text = this.makeSsmlMsg(text);
113 | 
114 |         this.synthesizing = true;
115 |         this.webSocket.send(text);
116 |     }
117 | 
118 |     /**
119 |      * 等待语音合成完成, 写入response
120 |      * @param response
121 |      */
122 |     public void writeAudioStream(HttpServletResponse response) {
123 |         long startTime = System.currentTimeMillis();
124 |         while (this.synthesizing) {
125 |             try {
126 |                 Thread.sleep(100);
127 |                 long time = System.currentTimeMillis() - startTime;
128 |                 if (time > 30 * 1000) {
129 |                     // 超时30s
130 |                     this.synthesizing = false;
131 |                 }
132 |             } catch (InterruptedException ex) {
133 |                 log.error("等待合成语音流异常", ex);
134 |             }
135 |         }
136 | 
137 |         byte[] bytes = buffer.readByteArray();
138 |         if (bytes.length > 0) {
139 |             response.setHeader("Content-Type", this.mime);
140 |             try (OutputStream out = response.getOutputStream()) {
141 |                 out.write(bytes, 0, bytes.length);
142 |             } catch (IOException ex) {
143 |                 log.error("语音流转储响应流失败: {}", ex.getMessage());
144 |                 response.setStatus(HttpServletResponse.SC_INTERNAL_SERVER_ERROR);
145 |             }
146 |         }
147 |         // 释放资源
148 |         this.buffer = null;
149 |     }
150 | 
151 |     /**
152 |      * 音频流写入缓存
153 |      * @param byteString
154 |      */
155 |     public void writeBuffer(ByteString byteString) {
156 |         this.buffer.write(byteString);
157 |     }
158 | 
159 |     /**
160 |      * websocket不可再使用, 丢弃
161 |      */
162 |     public void drop() {
163 |         this.available = false;
164 |         // 关闭websocket, 释放资源
165 |         this.webSocket.close(1000, null);
166 |         this.webSocket = null;
167 |     }
168 | 
169 |     /**
170 |      * 构造格式化消息
171 |      * @param text
172 |      * @return
173 |      */
174 |     private String makeSsmlMsg(String text) {
175 |         // 默认 "+0Hz"
176 |         int pitch = 0;
177 |         // 语速+, 默认 "+0%"
178 |         int rate = 0;
179 |         // 音量+, 默认 "+0%"
180 |         int volumn = 0;
181 |         String pitchStr = pitch >= 0 ? "+" + pitch + "Hz" : pitch + "Hz";
182 |         String rateStr = rate >= 0 ? "+" + rate + "%" : rate + "%";
183 |         String volumnStr = volumn >= 0 ? "+" + volumn + "%" : volumn + "%";
184 | 
185 |         String str = "X-RequestId:" + connectId() + "\r\n" +
186 |                 "Content-Type:application/ssml+xml\r\n" +
187 |                 "X-Timestamp:" + ttsConfig.getTime() + "Z\r\n" +
188 |                 "Path:ssml\r\n\r\n" +
189 |                 "<speak version=\"1.0\" xmlns=\"http://www.w3.org/2001/10/synthesis\" xmlns:mstts=\"https://www.w3.org/2001/mstts\" xml:lang=\"" + "en-US" + "\">" +
190 |                 "<voice  name=\"" + ttsConfig.getVoiceShortName() + "\">" +
191 |                 "<prosody pitch=\"" + pitchStr + "\" " +
192 |                 "rate =\"" + rateStr + "\" " +
193 |                 "volume=\"" + volumnStr + "\">"
194 |                 + text
195 |                 + "</prosody></voice></speak>";
196 |         if (log.isDebugEnabled()) {
197 |             log.debug("send:\n{}", str);
198 |         }
199 |         return str;
200 |     }
201 | 
202 |     /**
203 |      * get uuid
204 |      * @return
205 |      */
206 |     private String connectId() {
207 |         return UUID.randomUUID().toString().replace("-", "");
208 |     }
209 | 
210 |     /**
211 |      * 创建连接服务的websocket
212 |      * @return
213 |      */
214 |     private WebSocket createWs() {
215 |         String url = ttsConfig.getWssUrl() + "?TrustedClientToken=" + ttsConfig.getTrustedClientToken() +
216 |                 "&ConnectionId=" + connectId();
217 |         Request request = new Request.Builder()
218 |                 .url(url)
219 |                 .header("User-Agent", "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/92.0.4515.159 Safari/537.36 Edg/92.0.902.84")
220 |                 .header("Origin", "chrome-extension://jdiccldimpdaibmpdkjnbmckianbfold")
221 |                 .header("Pragma", "no-cache")
222 |                 .header("Cache-Control", "no-cache")
223 |                 .header("Accept-Encoding", "gzip, deflate, br")
224 |                 .header("Accept-Language", "zh-CN,zh;q=0.9,en;q=0.8,en-GB;q=0.7,en-US;q=0.6")
225 |                 .build();
226 |         // 可以使用匿名内部类
227 |         WebSocket webSocket = new OkHttpClient().newWebSocket(request, new TTSWebSocketListener(this));
228 |         String configMsg = ttsConfig.getSynthesizeConfig();
229 |         if (log.isDebugEnabled()) {
230 |             log.debug("send:\n{}", configMsg);
231 |         }
232 |         webSocket.send(configMsg);
233 |         return webSocket;
234 |     }
235 | 
236 |     /**
237 |      * 转义xml特殊字符
238 |      * @param str
239 |      * @return
240 |      */
241 |     private String escapeXmlChar(String str) {
242 |         str = str.replace("&", "&amp;");
243 |         str = str.replace(">", "&gt;");
244 |         str = str.replace("<", "&lt;");
245 |         return str;
246 |     }
247 | 
248 |     private int getSerialNum() {
249 |         int n = serialNum++;
250 |         if (serialNum == Integer.MAX_VALUE) {
251 |             serialNum = 1;
252 |         }
253 |         return n;
254 |     }
255 | }
256 | 


--------------------------------------------------------------------------------
/src/main/java/com/taoing/ttsserver/ws/TTSWebSocketListener.java:
--------------------------------------------------------------------------------
  1 | package com.taoing.ttsserver.ws;
  2 | 
  3 | import lombok.extern.slf4j.Slf4j;
  4 | import okhttp3.Response;
  5 | import okhttp3.WebSocket;
  6 | import okhttp3.WebSocketListener;
  7 | import okio.ByteString;
  8 | import org.jetbrains.annotations.NotNull;
  9 | import org.jetbrains.annotations.Nullable;
 10 | 
 11 | import java.nio.charset.StandardCharsets;
 12 | 
 13 | @Slf4j
 14 | public class TTSWebSocketListener extends WebSocketListener {
 15 | 
 16 |     /**
 17 |      * websocket client
 18 |      */
 19 |     private TTSWebSocket client;
 20 | 
 21 | 
 22 |     public TTSWebSocketListener() {
 23 |     }
 24 | 
 25 |     public TTSWebSocketListener(TTSWebSocket client) {
 26 |         this.client = client;
 27 |     }
 28 | 
 29 |     /**
 30 |      * websocket完全关闭
 31 |      * @param webSocket
 32 |      * @param code
 33 |      * @param reason
 34 |      */
 35 |     @Override
 36 |     public void onClosed(@NotNull WebSocket webSocket, int code, @NotNull String reason) {
 37 |         // 触发onFailure后, 再调用close方法, 也不会再触发该方法
 38 |         log.info("onClosed: {}-{}", code, reason);
 39 |         if (log.isDebugEnabled()) {
 40 |             log.debug("websocket完全关闭: {}", client.toString());
 41 |         }
 42 |         client.setAvailable(false);
 43 |     }
 44 | 
 45 |     /**
 46 |      * 服务端关闭连接, 会触发该方法
 47 |      * @param webSocket
 48 |      * @param code
 49 |      * @param reason
 50 |      */
 51 |     @Override
 52 |     public void onClosing(@NotNull WebSocket webSocket, int code, @NotNull String reason) {
 53 |         log.info("onClosing: {}-{}", code, reason);
 54 |         if (log.isDebugEnabled()) {
 55 |             log.debug("websocket服务端关闭: {}", client.toString());
 56 |         }
 57 |         // 设置websocket不可用, 进一步关闭该websocket以释放资源
 58 |         client.setAvailable(false);
 59 |     }
 60 | 
 61 |     @Override
 62 |     public void onFailure(@NotNull WebSocket webSocket, @NotNull Throwable t, @Nullable Response response) {
 63 |         // websocket连接空闲30s后, 服务端主动断开连接, 触发了该方法, 后续该websocket不再可用
 64 |         client.drop();
 65 | 
 66 |         String message = t.getMessage();
 67 |         if (response != null) {
 68 |             message += "-" + response.message();
 69 |         }
 70 |         log.warn("websocket与服务端中断连接: {}, onFailure: {}", client.toString(), message);
 71 |     }
 72 | 
 73 |     @Override
 74 |     public void onMessage(@NotNull WebSocket webSocket, @NotNull String text) {
 75 |         if (log.isDebugEnabled()) {
 76 |             log.debug("recv:\n{}", text);
 77 |         }
 78 |         String startTag = "turn.start";
 79 |         String endTag = "turn.end";
 80 |         int startIndex = text.indexOf(startTag);
 81 |         int endIndex = text.indexOf(endTag);
 82 |         // 生成开始
 83 |         if (startIndex != -1) {
 84 |             if (log.isDebugEnabled()) {
 85 |                 log.debug("turn.start");
 86 |             }
 87 |         } else if (endIndex != -1) {
 88 |             // 生成结束
 89 |             client.setSynthesizing(false);
 90 |             if (log.isDebugEnabled()) {
 91 |                 log.debug("turn.end");
 92 |             }
 93 |         }
 94 |     }
 95 | 
 96 |     @Override
 97 |     public void onMessage(@NotNull WebSocket webSocket, @NotNull ByteString bytes) {
 98 |         if (log.isDebugEnabled()) {
 99 |             log.debug("recv:\n{}", bytes.utf8());
100 |         }
101 |         // 音频数据流标志头
102 |         String audioTag = "Path:audio\r\n";
103 |         String startTag = "Content-Type:";
104 |         String endTag = "\r\nX-StreamId";
105 | 
106 |         int audioIndex = bytes.indexOf(audioTag.getBytes(StandardCharsets.UTF_8));
107 |         int startIndex = bytes.indexOf(startTag.getBytes(StandardCharsets.UTF_8));
108 |         int endIndex = bytes.indexOf(endTag.getBytes(StandardCharsets.UTF_8));
109 |         if (audioIndex != -1) {
110 |             audioIndex += audioTag.length();
111 |             if (client.getMime() == null) {
112 |                 client.setMime(bytes.substring(startIndex + startTag.length(), endIndex).utf8());
113 |             }
114 |             if ("audio/x-wav".equals(client.getMime()) && bytes.indexOf("RIFF".getBytes(StandardCharsets.UTF_8)) != -1) {
115 |                 // 去除WAV文件的文件头，解决播放开头时的杂音
116 |                 audioIndex += 44;
117 |             }
118 |             client.writeBuffer(bytes.substring(audioIndex));
119 |         }
120 |     }
121 | 
122 |     @Override
123 |     public void onOpen(@NotNull WebSocket webSocket, @NotNull Response response) {
124 |         if (log.isDebugEnabled()) {
125 |             log.debug("微软tts websocket服务已连接: {}", client.toString());
126 |         }
127 |     }
128 | }
129 | 


--------------------------------------------------------------------------------
/src/main/java/com/taoing/ttsserver/ws/WebSocketManager.java:
--------------------------------------------------------------------------------
 1 | package com.taoing.ttsserver.ws;
 2 | 
 3 | import lombok.Getter;
 4 | import lombok.Setter;
 5 | import lombok.extern.slf4j.Slf4j;
 6 | import org.springframework.beans.factory.annotation.Autowired;
 7 | import org.springframework.stereotype.Component;
 8 | 
 9 | import java.util.HashMap;
10 | import java.util.Map;
11 | 
12 | /**
13 |  * websocket client 管理
14 |  */
15 | @Slf4j
16 | @Component
17 | @Getter
18 | @Setter
19 | public class WebSocketManager {
20 |     /**
21 |      * 可用websocket client识别编码, 最多3个
22 |      */
23 |     public static final String[] AVAILABLE_CODES = {"xxttsa01", "xxttsa02", "xxttsa03"};
24 | 
25 |     @Autowired
26 |     private TTSConfig ttsConfig;
27 | 
28 |     private Map<String, TTSWebSocket> webSocketMap;
29 | 
30 |     public WebSocketManager() {
31 |         webSocketMap = new HashMap<>();
32 |     }
33 | 
34 |     public TTSWebSocket getClient(String code) {
35 |         boolean isMatch = false;
36 |         for (String item : AVAILABLE_CODES) {
37 |             if (item.equals(code)) {
38 |                 isMatch = true;
39 |                 break;
40 |             }
41 |         }
42 |         if (!isMatch) {
43 |             return null;
44 |         }
45 |         TTSWebSocket webSocket = webSocketMap.get(code);
46 |         if (webSocket == null || !webSocket.isAvailable()) {
47 |             webSocket = new TTSWebSocket(code, ttsConfig);
48 |             webSocketMap.put(code, webSocket);
49 |         }
50 |         return webSocket;
51 |     }
52 | }
53 | 


--------------------------------------------------------------------------------
/src/main/resources/application-dev.yml:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/taoing/tts-server/616d91fde597c1ab5bfae3552cb03613306e3df9/src/main/resources/application-dev.yml


--------------------------------------------------------------------------------
/src/main/resources/application-pro.yml:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/taoing/tts-server/616d91fde597c1ab5bfae3552cb03613306e3df9/src/main/resources/application-pro.yml


--------------------------------------------------------------------------------
/src/main/resources/application-test.yml:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/taoing/tts-server/616d91fde597c1ab5bfae3552cb03613306e3df9/src/main/resources/application-test.yml


--------------------------------------------------------------------------------
/src/main/resources/application.yml:
--------------------------------------------------------------------------------
 1 | server:
 2 |   port: 9027
 3 | 
 4 | spring:
 5 |   application:
 6 |     name: tts-server
 7 |   profiles:
 8 |     active: dev
 9 |   jackson:
10 |     time-zone: GMT+8
11 |     date-format: yyyy-MM-dd HH:mm:ss
12 |     deserialization:
13 |       # 允许对象忽略json中不存在的属性
14 |       fail_on_unknown_properties: false
15 | 
16 | logging:
17 |   config: classpath:logback-spring.xml
18 | 
19 | # 微软edge浏览器tts语音朗读配置
20 | tts:
21 |   edge:
22 |     # https://speech.platform.bing.com/consumer/speech/synthesize/readaloud/voices/list?trustedclienttoken=6A5AA1D4EAFF4E9FB37E23D68491D6F4
23 |     # 可查看可用tts列表
24 |     voice-list-url: https://speech.platform.bing.com/consumer/speech/synthesize/readaloud/voices/list # 微软可用声音tts列表查询url
25 |     wss-url: wss://speech.platform.bing.com/consumer/speech/synthesize/readaloud/edge/v1 # 语音合成websocket服务
26 |     trusted-client-token: 6A5AA1D4EAFF4E9FB37E23D68491D6F4 # token
27 |     voice-name: Microsoft Server Speech Text to Speech Voice (zh-CN, XiaoxiaoNeural) # 默认使用的声音tts(微软晓晓)
28 |     voice-short-name: zh-CN-XiaoxiaoNeural # 默认使用的声音tts(微软晓晓简写)
29 |     codec: audio-24khz-48kbitrate-mono-mp3 # 声音编码
30 |     local: zh-CN # 区域
31 |     sentenceBoundaryEnabled: false # 可以试试true或false的效果(感知不到差别)
32 |     wordBoundaryEnabled: false # 可以试试true或false的效果(感知不到差别)
33 | 


--------------------------------------------------------------------------------
/src/main/resources/logback-spring.xml:
--------------------------------------------------------------------------------
 1 | <?xml version="1.0" encoding="UTF-8"?>
 2 | <configuration>
 3 |     <!-- ConsoleAppender：把日志输出到控制台 -->
 4 |     <appender name="console" class="ch.qos.logback.core.ConsoleAppender">
 5 |         <encoder>
 6 |             <pattern>%d %p (%file:%line\)- %m%n</pattern>
 7 |             <!-- 控制台也要使用UTF-8，不要使用GBK，否则会中文乱码 -->
 8 |             <charset>UTF-8</charset>
 9 |         </encoder>
10 |     </appender>
11 | 
12 |     <!-- RollingFileAppender：滚动记录文件，先将日志记录到指定文件，当符合某个条件时，将日志记录到其他文件 -->
13 |     <!-- 以下的大概意思是：1.先按日期存日志，日期变了，将前一天的日志文件名重命名为XXX%日期%索引，新的日志仍然是demo.log -->
14 |     <!--             2.如果日期没有发生变化，但是当前日志的文件大小超过1KB时，对当前日志进行分割 重命名-->
15 |     <appender name="filelog" class="ch.qos.logback.core.rolling.RollingFileAppender">
16 |         <File>./logs/tts-server.log</File>
17 |         <!-- rollingPolicy:当发生滚动时，决定 RollingFileAppender 的行为，涉及文件移动和重命名。 -->
18 |         <!-- TimeBasedRollingPolicy： 最常用的滚动策略，它根据时间来制定滚动策略，既负责滚动也负责出发滚动 -->
19 |         <rollingPolicy class="ch.qos.logback.core.rolling.TimeBasedRollingPolicy">
20 |             <!-- 活动文件的名字会根据fileNamePattern的值，每隔一段时间改变一次 -->
21 |             <!-- 文件名：log/sdk-ucenter.2017-12-05.0.log -->
22 |             <fileNamePattern>./logs/tts-server.%d.%i.log</fileNamePattern>
23 |             <!-- 每产生一个日志文件，该日志文件的保存期限为30天 -->
24 |             <maxHistory>30</maxHistory>
25 |             <timeBasedFileNamingAndTriggeringPolicy class="ch.qos.logback.core.rolling.SizeAndTimeBasedFNATP">
26 |                 <!-- maxFileSize:这是活动文件的大小，默认值是10MB，测试时可改成1KB看效果 -->
27 |                 <maxFileSize>10MB</maxFileSize>
28 |             </timeBasedFileNamingAndTriggeringPolicy>
29 |         </rollingPolicy>
30 |         <encoder>
31 |             <!-- pattern节点，用来设置日志的输入格式 -->
32 |             <pattern>
33 |                 %d %p (%file:%line\)- %m%n
34 |             </pattern>
35 |             <!-- 记录日志的编码:此处设置字符集 - -->
36 |             <charset>UTF-8</charset>
37 |         </encoder>
38 |     </appender>
39 | 
40 |     <!-- 日志输出级别 -->
41 |     <root level="INFO">
42 |         <appender-ref ref="console" />
43 |         <appender-ref ref="filelog" />
44 |     </root>
45 | 
46 | <!--    <logger name="com.taoing" level="DEBUG" />-->
47 | <!--    <logger name="com.taoing.ttsserver.controller" level="DEBUG" />-->
48 | 
49 |     <springProfile name="test">
50 |         <root level="ERROR">
51 |             <appender-ref ref="filelog" />
52 |         </root>
53 |         <logger name="com.taoing" level="INFO" />
54 |     </springProfile>
55 | 
56 |     <springProfile name="pro">
57 |         <root level="ERROR">
58 |             <appender-ref ref="filelog" />
59 |         </root>
60 |         <logger name="com.taoing" level="INFO" />
61 |     </springProfile>
62 | </configuration>


--------------------------------------------------------------------------------
/src/test/java/com/taoing/ttsserver/TtsServerApplicationTests.java:
--------------------------------------------------------------------------------
 1 | package com.taoing.ttsserver;
 2 | 
 3 | import org.junit.jupiter.api.Test;
 4 | import org.springframework.boot.test.context.SpringBootTest;
 5 | 
 6 | @SpringBootTest
 7 | class TtsServerApplicationTests {
 8 | 
 9 |     @Test
10 |     void contextLoads() {
11 |     }
12 | 
13 | }
14 | 


--------------------------------------------------------------------------------
/startup.sh:
--------------------------------------------------------------------------------
 1 | #!/bin/bash
 2 | #发布tts-server, jar包上传到 ./upload 目录
 3 | 
 4 | set -x
 5 | set -e
 6 | 
 7 | uploadDir=./upload
 8 | jarfile=tts-server-1.0.0.jar
 9 | 
10 | #kill进程
11 | if [[ $(ps -ef | grep "${jarfile}" | grep -v "grep") != "" ]]; then
12 |   ps -ef | grep "${jarfile}" | grep -v "grep" | awk '{print $2}' | xargs kill
13 | fi
14 | sleep 2
15 | if [ -f ./$jarfile ]; then
16 | 	rm -rf ./$jarfile
17 | fi
18 | cp -f $uploadDir/$jarfile ./$jarfile
19 | 
20 | nohup java -jar -Dspring.profiles.active=test $jarfile > /dev/null 2>&1 &
21 | echo "${jarfile} 发布完成"
22 | 


--------------------------------------------------------------------------------