智能语音交互--声音事件检测Quick Start

使用前提与环境准备：服务开通并购买

• 智能语音交互：开通并购买地址
• 控制台获取Token
• 控制台创建项目，获得项目appkey

操作步骤：

1.添加pom依赖

<dependency>
    <groupId>com.alibaba.nls</groupId>
    <artifactId>nls-sdk-recognizer</artifactId>
    <version>2.2.1</version>
</dependency>
<dependency>
  <groupId>com.alibaba.nls</groupId>
  <artifactId>nls-sdk-common</artifactId>
  <version>2.2.1</version>
</dependency>
<dependency>
  <groupId>com.alibaba.nls</groupId>
  <artifactId>nls-sdk-request</artifactId>
  <version>2.2.1</version>
</dependency>

        

          
        
        
        

          

          AI 代码解读
        
      
      

2.Code Sample

import com.alibaba.nls.client.protocol.NlsClient;
import com.alibaba.nls.client.protocol.commonrequest.CommonRequest;
import com.alibaba.nls.client.protocol.commonrequest.CommonRequestListener;
import com.alibaba.nls.client.protocol.commonrequest.CommonRequestResponse;
import java.io.InputStream;
import java.util.Arrays;
//声音事件检测
public class AudioEventDetectionDemo {

    public static final String TOKEN = "XXXXXXXXXX";

    public static final String APPKEY = "XXXXXXXXXX";

    private static final String NAMESPACE = "AudioEventDetection";

    private static final String URL = "wss://nls-gateway.cn-shanghai.aliyuncs.com/ws/v1";

    private static final int SAMPLE_RATE = 16000;

    private static final int CHUNK_DURATION = 200;

    public static void main(String[] args) throws Exception {
        NlsClient client = new NlsClient(URL, TOKEN);

        //InputStream stream = AudioEventDetectionDemo.class.getResourceAsStream("/test.pcm");
        FileInputStream stream = new FileInputStream("D:\\download\\test.pcm");

        CommonRequestListener listener = getListener();
        CommonRequest request = new CommonRequest(client, listener, NAMESPACE);
        request.setAppKey(APPKEY);
        request.addCustomedParam("format", "pcm");
        request.addCustomedParam("sample_rate", 16000);
        request.start();
        int chunkSize = SAMPLE_RATE * 2 / 1000 * CHUNK_DURATION;
        byte[] data = new byte[chunkSize];
        while (true) {
            int len = stream.read(data);
            if (len < 0) {
                break;
            }
            if (len > 0) {
                request.send(Arrays.copyOf(data, len));
            }
            Thread.sleep(CHUNK_DURATION / 10);
        }
        request.stop();

        client.shutdown();
    }

    private static CommonRequestListener getListener() {
        CommonRequestListener listener = new CommonRequestListener() {
            @Override
            public void onStarted(CommonRequestResponse response) {
                System.out.println(
                        "onStarted, taskId: " + response.getTaskId() + ", header: " + response.header + ", payload: "
                                + response.payload);
            }

            @Override
            public void onEvent(CommonRequestResponse response) {
                System.out.println(
                        "onEvent, taskId: " + response.getTaskId() + ", header: " + response.header + ", payload: "
                                + response.payload);
            }

            @Override
            public void onStopped(CommonRequestResponse response) {
                System.out.println(
                        "onStopped, taskId: " + response.getTaskId() + ", header: " + response.header + ", payload: "
                                + response.payload);
            }

            @Override
            public void onFailed(CommonRequestResponse response) {
                System.out.println(
                        "onFailed, taskId: " + response.getTaskId() + ", header: " + response.header + ", payload: "
                                + response.payload);
            }
        };
        return listener;
    }
}

        

          
        
        
        

          

          AI 代码解读
        
      
      

3.测试结果

onFailed, taskId: f0d669851a1a40e8804d1e11651b756c, header: {namespace=Default, name=TaskFailed, message_id=80455ce2482748b1b85260a4634f5c93, task_id=f0d669851a1a40e8804d1e11651b756c, status_text=Gateway:FREE_TRIAL_EXPIRED:The free trial has expired!, status=40000010}, payload: {}

        

          
        
        
        

          

          AI 代码解读
        
      
      

• 目前由于语音分析服务并不支持免费试用版，需要控制台开通商用版才可正常试用

4.升级商用后再次测试结果

onEvent, taskId: f06635cda4624969850a314ab99fc6db, header: {namespace=AudioEventDetection, name=TaskResult, message_id=de21af540281426e8c6cd573fb92c15e, task_id=f06635cda4624969850a314ab99fc6db, status_text=Gateway:SUCCESS:Success., status=20000000}, payload: {result=[{"start_time":0,"event_type":"Music","confidence":0.8432333469390869,"end_time":1970}], number=1}

        

          
        
        
        

          

          AI 代码解读
        
      
      

更多参考

开发指南
声音事件检测