鸿蒙系统音频开发

牵扯的能力

1. 录音
2. 权限声明、申请
3. 用户文件管理

音频录制方式

PCM全称Pulse-Code Modulation，翻译为脉冲调制编码，是一种用数字表示模拟信号的方法

开发步骤

代码实现基于鸿蒙API 9 Release，其他版本可能存在接口差异。需要对录制的音频进行实时处理的（比如语音交互）需要采用AudioCapturer来实现音频录制，这里也介绍一下此种方式的实现。

权限的声明和申请

应用需要调用麦克风来录制音频，该行为属于隐私敏感行为，需要在调用麦克风前申请权限

1. 申明麦克风使用权限

   在module.json5配置文件的requestPermissions标签中声明权限

   {
     "module": {
       "name": "entry",
       "type": "entry",
       "description": "$string:module_desc",
       "mainElement": "EntryAbility",
       ......
       "abilities": [
         {
           "name": "EntryAbility",
           "srcEntry": "./ets/entryability/EntryAbility.ts",
           ......
         }
       ],
       "requestPermissions": [
         {
           "name": "ohos.permission.MICROPHONE",
           "reason": "$string:permission_microphone_reason",
           "usedScene": {
             "when": "always"
           }
         }
       ]
     }
   }
   

   其中name字段的ohos.permission.MICROPHONE为麦克风使用权限名，reason字段标明申请权限的原因，usedScene标明权限使用的场景

2. 向用户申请麦克风使用权限

   首先校验当前是否已经授权，如果已经授权可直接发起录音

   // 申明需要申请的权限，这里只需麦克风权限
   const PERMISSIONS: Array<Permissions> = ['ohos.permission.MICROPHONE']
   
   // 检查麦克风权限是否授权
   private async checkMicRophonePermission(): Promise<boolean> {
     let grantStatus: abilityAccessCtrl.GrantStatus = await this.checkPermissionGrant(PERMISSIONS[0])
     if (grantStatus === abilityAccessCtrl.GrantStatus.PERMISSION_GRANTED) {
       return true
     } else {
       return false
     }
   }
   
   // 检查参数中的权限是否授权
   private async checkPermissionGrant(permission: Permissions): Promise<abilityAccessCtrl.GrantStatus> {
     let atManager: abilityAccessCtrl.AtManager = abilityAccessCtrl.createAtManager();
     let grantStatus: abilityAccessCtrl.GrantStatus = abilityAccessCtrl.GrantStatus.PERMISSION_DENIED
   
     // 获取应用程序accessTokenID
     let tokenId: number = 0
     try {
       let bundleInfo: bundleManager.BundleInfo = await bundleManager.getBundleInfoForSelf(bundleManager.BundleFlag.GET_BUNDLE_INFO_WITH_APPLICATION)
       let appInfo: bundleManager.ApplicationInfo = bundleInfo.appInfo
       tokenId = appInfo.accessTokenId
     } catch (error) {
       // const err = error as BusinessError
       Logger.error(TAG, `Fialed to get bundle info for self. COde is ${error.code}, message is ${error.message}`)
     }
   
     // 校验应用是否被授权
     try {
       grantStatus = await atManager.checkAccessToken(tokenId, permission)
     } catch (error) {
       Logger.error(TAG, `Failed to check access token, COde is ${error.code}}, messsage is ${error.message}}`)
     }
     return grantStatus
   }
   

   如果判断未被授权需要主动向用户申请权限，在用户授权后发起录音功能

   private reqPermissionsFromUser() {
     let context = getContext(this) as common.UIAbilityContext;
     let atManager: abilityAccessCtrl.AtManager = abilityAccessCtrl.createAtManager()
     try {
       atManager.requestPermissionsFromUser(context, PERMISSIONS).then((data) => {
         let grantStatus: Array<number> = data.authResults
         let length: number = grantStatus.length
         for (let i = 0; i < length; i++) {
           if (grantStatus[i] === 0) {
             promptAction.showToast({ message: '授予录音权限成功' })
             // 发起录音
             this.startAudio()
           } else {
             promptAction.showToast({ message: '授予录音权限失败' })
             return;
           }
         }
       })
     } catch (error) {
       // const err = error as BusinessError
       Logger.error(TAG, `Failed to request permissions from user. Code is ${error.code}, message is ${error.message}`);
     }
   }
   

保存文件的获取

这里录音的实现是将实时录制的音频保存到文件中，如果采用“应用文件”的方式保存可直接采用如下的方式获取文件

 // 获取应用文件路径
 let context = getContext(this) as common.UIAbilityContext
 let path = context.cacheDir;
 // 新建并打开文件
 let filePath = path + '/temp.pcm';
 let file: fs.File = fs.openSync(filePath, fs.OpenMode.READ_WRITE | fs.OpenMode.CREATE);

如需要保存为“用户文件”，可通过拉起“文件选择器应用”选择目录来实现。

private async createAudioFile(): Promise<string> {
	// 创建音频保存选项实例
  const audioSaveOptions = new picker.AudioSaveOptions();
  // 设定保存文件名
  audioSaveOptions.newFileNames = ['test.pcm'];
  // 采用音频选择器
  const audioViewPicker = new picker.AudioViewPicker();
  // 发起目标文件夹选择并获取文件的uri
  let uris = await audioViewPicker.save(audioSaveOptions)
  if (uris !== undefined && uris.length > 0) {
    Logger.info(TAG, 'audioViewPicker.save to file succeed and uri is:' + uris[0]);
    return uris[0]
  } else {
    return ''
  }
}

音频的录制

下图展示了AudioCapturer的状态变化，在创建实例后，调用对应的方法可以进入指定的状态实现对应的行为

1. 配置音频采集参数

   let audioSteamInfo: audio.AudioStreamInfo = {
     samplingRate: audio.AudioSamplingRate.SAMPLE_RATE_16000, // 采样率
     channels: audio.AudioChannel.CHANNEL_1, // 通道
     sampleFormat: audio.AudioSampleFormat.SAMPLE_FORMAT_S16LE, // 采样格式
     encodingType: audio.AudioEncodingType.ENCODING_TYPE_RAW // 编码格式
   }
   
   let audioCaptureInfo: audio.AudioCapturerInfo = {
     source: audio.SourceType.SOURCE_TYPE_MIC, // 音源类型
     capturerFlags: 0 // 音频采集器标志，0代表普通音频采集器
   }
   
   let audioCaptureOptions: audio.AudioCapturerOptions = {
     streamInfo: audioSteamInfo,
     capturerInfo: audioCaptureInfo
   }
   

2. 创建AudioCapturer实例

   try {
     this.audioCapturer = await audio.createAudioCapturer(audioCaptureOptions)
     Logger.info(TAG, 'Invoke createAudioCapturer succeeded.')
     if (this.audioCapturer !== undefined) {
   	  // 创建成功可以监听音频状态变更
       this.audioCapturer.on('stateChange', (state: audio.AudioState) => {
         Logger.info(TAG, `current state is ${state}`);
       })
     }
   } catch (err) {
     Logger.error(TAG, `Invoke createAudioCapturer failed, code is ${err.code}, message is ${err.message}`)
   }
   

3. 调用start()方法进入running状态，开始录制音频

   if (this.audioCapturer !== undefined) {
     let stateGroup = [audio.AudioState.STATE_PREPARED, audio.AudioState.STATE_PAUSED, audio.AudioState.STATE_STOPPED];
     if (stateGroup.indexOf(this.audioCapturer.state.valueOf()) === -1) {
       Logger.error(TAG, `start failed`);
       return;
     }
     this.audioCapturer.start((err: Error) => {
       if (err) {
         Logger.error(TAG, `Capturer start failed, code is ${err.name}, message is ${err.message}`);
       } else {
         Logger.info(TAG, 'Capturer start success.');
         // 保存录音的音频到文件
         this.startSaveAudio()
       }
     })
   }
   

4. 在录音状态循环读取数据并保存到文件中，直接录制结束

   class Options {
     offset?: number;
     length?: number;
   }
   
   // 获取用于保存录音的合理最小缓冲区大小
   let bufferSize = await this.audioCapturer.getBufferSize()
   
   while (this.audioCapturer.state.valueOf() === audio.AudioState.STATE_RUNNING) {
     try {
       let buffer = await this.audioCapturer.read(bufferSize, true)
       if (buffer !== undefined) {
         Logger.info(TAG, 'buffer read successfully');
         let options: Options = {
           offset: this.audioSize,
           length: buffer.byteLength
         }
         fs.writeSync(this.audioFile.fd, buffer, options);
         this.audioSize += buffer.byteLength;
       }
     } catch (err) {
       Logger.info(TAG, `save audio file error ${err}`);
     }
   }
   // 录制完成关闭文件
   fs.close(this.audioFile);
   

5. 停止录制

   if (this.audioCapturer !== undefined) {
     // 只有采集器状态为STATE_RUNNING或STATE_PAUSED的时候才可以停止
     if (this.audioCapturer.state.valueOf() !== audio.AudioState.STATE_RUNNING && this.audioCapturer.state.valueOf() !== audio.AudioState.STATE_PAUSED) {
       Logger.info(TAG, 'Capturer is not running or paused');
       return
     }
     this.audioCapturer.stop()
   }
   

6. 销毁实例，释放资源

   if (this.audioCapturer !== undefined) {
     // 采集器状态不是STATE_RELEASED或STATE_NEW状态，才能release
     if (this.audioCapturer.state.valueOf() === audio.AudioState.STATE_RELEASED || this.audioCapturer.state.valueOf() === audio.AudioState.STATE_NEW) {
       Logger.info(TAG, 'Capturer already released');
       return;
     }
     this.audioCapturer.release((err: Error) => {
       if (err) {
         Logger.error(TAG, 'Capturer release failed.');
       } else {
         Logger.info(TAG, 'Capturer release success.');
         this.audioCapturer = undefined;
       }
     });
   }
   

从总体来看鸿蒙录音接口基本上参照的Android定义，在细节上有部分优化（比如获取最小缓冲区大小方式），基于JavaScript 的语言特性不用启动单独的线程使用Promise异步调用即可获取保存录音数据。