版權(quán)聲明：本文為博主原創(chuàng)文章，未經(jīng)博主同意不得轉(zhuǎn)載。 https://blog.csdn.net/zpf8861/article/details/30229039

百度Android語音識別SDK分在線與離線兩種。這篇文章介紹在線SDK的用法。

? ? ?在線SDK是以JAR包和動態(tài)鏈接庫形式公布和使用。能夠從百度開放云平臺站點(diǎn)中下載SDK及使用說明文檔。

http://developer.baidu.com/wiki/index.php?title=docs/cplat/media/voice

? ? ?完畢語音SDK的集成分下面幾步。本文將一步步介紹SDK集成方法。

? ? ??

?1、注冊開放開放平臺 ??

?點(diǎn)擊管理控制臺。選擇移動應(yīng)用管理

?選擇創(chuàng)建應(yīng)用，填寫應(yīng)用名稱

? ?<img src="https://img-blog.csdn.net/20140529141311875?

watermark/2/text/aHR0cDovL2Jsb2cuY3Nkbi5uZXQvVm9pY2VGYW5z/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70/gravity/Center" alt="" />???

<img src="https://img-blog.csdn.net/20140529141506890?

watermark/2/text/aHR0cDovL2Jsb2cuY3Nkbi5uZXQvVm9pY2VGYW5z/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70/gravity/Center" alt="" />

能夠看到右上角有ID、API KEY、Secret KEY。點(diǎn)擊能夠復(fù)制其內(nèi)容，保存這些字符串，在使用語音SDK時(shí)會用到。

<img src="https://img-blog.csdn.net/20140529141527468?

watermark/2/text/aHR0cDovL2Jsb2cuY3Nkbi5uZXQvVm9pY2VGYW5z/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70/gravity/Center" alt="" />

2、申請開啟語音識別服務(wù) 。選擇媒體云---語音識別，點(diǎn)擊申請開啟服務(wù)，填寫理由。

等待對接成功

3、使用語音識別SDK前的準(zhǔn)備

之前準(zhǔn)備了SDK開發(fā)包以及ID、API KEY、Secret KEY。

首先將開發(fā)包中的lib中的庫加入到project中

? ? ? ? 聲明權(quán)限

[java]?view
plaincopy

?<uses-permission?android:name="android.permission.RECORD_AUDIO"?/>??
<uses-permission?android:name="android.permission.ACCESS_NETWORK_STATE"?/>??
<uses-permission?android:name="android.permission.WRITE_EXTERNAL_STORAGE"?/>??
<uses-permission?android:name="android.permission.INTERNET"?/>??
<uses-permission?android:name="android.permission.READ_PHONE_STATE"?/>??
<uses-permission?android:name="android.permission.WAKE_LOCK"?/>??
<uses-permission?android:name="android.permission.ACCESS_FINE_LOCATION"?/>??
<uses-permission?android:name="android.permission.ACCESS_COARSE_LOCATION"?/>??
<uses-permission?android:name="android.permission.ACCESS_WIFI_STATE"?/></span>??

???4、語音識別

SDK有兩種實(shí)現(xiàn)語音識別的方式，一種是直接使用SDK中的語音識別控件。一種是使用SDK中的語音識別服務(wù)。

語音識別控件方式

語音識別控件BaiduASRDigitalDialog。提供了整套語音交互、提示音、音量反饋、動效反饋。開發(fā)人員初始化一個(gè)BaiduASRDigitalDialog對象，并設(shè)置相關(guān)參數(shù)及結(jié)果回調(diào)，調(diào)用Show()方法就能夠彈出對話框開始識別。識別結(jié)束后會在回調(diào)中得到識別結(jié)果。

[java]?view
plaincopy

if?(mDialog?==?null?||?mCurrentTheme?!=?Config.DIALOG_THEME)?{??
????????????????????mCurrentTheme?=?Config.DIALOG_THEME;??
????????????????????if?(mDialog?!=?null)?{??
????????????????????????mDialog.dismiss();??
????????????????????}??
????????????????????Bundle?params?=?new?Bundle();??
????????????????????params.putString(BaiduASRDigitalDialog.PARAM_API_KEY,?Constants.API_KEY);??
????????????????????params.putString(BaiduASRDigitalDialog.PARAM_SECRET_KEY,?Constants.SECRET_KEY);??
????????????????????params.putInt(BaiduASRDigitalDialog.PARAM_DIALOG_THEME,?Config.DIALOG_THEME);??
????????????????????mDialog?=?new?BaiduASRDigitalDialog(this,?params);??
????????????????????mDialog.setDialogRecognitionListener(mRecognitionListener);??
????????????????}??
????????????????mDialog.getParams().putInt(BaiduASRDigitalDialog.PARAM_PROP,?Config.CURRENT_PROP);??
????????????????mDialog.getParams().putString(BaiduASRDigitalDialog.PARAM_LANGUAGE,??
????????????????????????Config.getCurrentLanguage());??
????????????????mDialog.show();??

識別對話框支持的參數(shù)定義在BaiduASRDigitalDialog中以PARAM_前綴的常量。列表例如以下：

設(shè)置回調(diào)方法。處理返回的結(jié)果

[java]?view
plaincopy

mRecognitionListener?=?new?DialogRecognitionListener()?{??
??
???????????@Override??
???????????public?void?onResults(Bundle?results)?{??
???????????????ArrayList<String>?rs?=?results?!=?null???results??
???????????????????????.getStringArrayList(RESULTS_RECOGNITION)?:?null;??
???????????????if?(rs?!=?null?&&?rs.size()?>?0)?{??
???????????????????mResult.setText(rs.get(0));??
???????????????}??
??
???????????}??
???????};??

API方式

首先須要配置語音識別引擎ASREngine的參數(shù)VoiceRecognitionConfig

[java]?view
plaincopy

VoiceRecognitionConfig?config?=?new?VoiceRecognitionConfig();??
????????????????config.setProp(Config.CURRENT_PROP);??
????????????????config.setLanguage(Config.getCurrentLanguage());??
????????????????config.enableVoicePower(Config.SHOW_VOL);?//?音量反饋。??
????????????????if?(Config.PLAY_START_SOUND)?{??
????????????????????config.enableBeginSoundEffect(R.raw.bdspeech_recognition_start);?//?設(shè)置識別開始提示音??
????????????????}??
????????????????if?(Config.PLAY_END_SOUND)?{??
????????????????????config.enableEndSoundEffect(R.raw.bdspeech_speech_end);?//?設(shè)置識別結(jié)束提示音??
????????????????}??
????????????????config.setSampleRate(VoiceRecognitionConfig.SAMPLE_RATE_8K);?//?設(shè)置採樣率,須要與外部音頻一致??

然后啟動識別

[java]?view
plaincopy

int?code?=?mASREngine.startVoiceRecognition(mListener,?config);??

當(dāng)中mListener是識別過程的回調(diào)，須要對當(dāng)中的方法進(jìn)行實(shí)現(xiàn)

[java]?view
plaincopy

/**?
????*?重寫用于處理語音識別回調(diào)的監(jiān)聽器?
????*/??
???class?MyVoiceRecogListener?implements?VoiceClientStatusChangeListener?{??
??
???????@Override??
???????public?void?onClientStatusChange(int?status,?Object?obj)?{??
???????????switch?(status)?{??
???????????//?語音識別實(shí)際開始，這是真正開始識別的時(shí)間點(diǎn)。需在界面提示用戶說話。??
???????????????case?VoiceRecognitionClient.CLIENT_STATUS_START_RECORDING:??
???????????????????isRecognition?=?true;??
???????????????????mHandler.removeCallbacks(mUpdateVolume);??
???????????????????mHandler.postDelayed(mUpdateVolume,?POWER_UPDATE_INTERVAL);??
???????????????????mControlPanel.statusChange(ControlPanelFragment.STATUS_RECORDING_START);??
???????????????????break;??
???????????????case?VoiceRecognitionClient.CLIENT_STATUS_SPEECH_START:?//?檢測到語音起點(diǎn)??
???????????????????mControlPanel.statusChange(ControlPanelFragment.STATUS_SPEECH_START);??
???????????????????break;??
???????????????//?已經(jīng)檢測到語音終點(diǎn)，等待網(wǎng)絡(luò)返回??
???????????????case?VoiceRecognitionClient.CLIENT_STATUS_SPEECH_END:??
???????????????????mControlPanel.statusChange(ControlPanelFragment.STATUS_SPEECH_END);??
???????????????????break;??
???????????????//?語音識別完畢，顯示obj中的結(jié)果??
???????????????case?VoiceRecognitionClient.CLIENT_STATUS_FINISH:??
???????????????????mControlPanel.statusChange(ControlPanelFragment.STATUS_FINISH);??
???????????????????isRecognition?=?false;??
???????????????????updateRecognitionResult(obj);??
???????????????????break;??
???????????????//?處理連續(xù)上屏??
???????????????case?VoiceRecognitionClient.CLIENT_STATUS_UPDATE_RESULTS:??
???????????????????updateRecognitionResult(obj);??
???????????????????break;??
???????????????//?用戶取消??
???????????????case?VoiceRecognitionClient.CLIENT_STATUS_USER_CANCELED:??
???????????????????mControlPanel.statusChange(ControlPanelFragment.STATUS_FINISH);??
???????????????????isRecognition?=?false;??
???????????????????break;??
???????????????default:??
???????????????????break;??
???????????}??
??
???????}??
??
???????@Override??
???????public?void?onError(int?errorType,?int?errorCode)?{??
???????????isRecognition?=?false;??
???????????mResult.setText(getString(R.string.error_occur,?Integer.toHexString(errorCode)));??
???????????mControlPanel.statusChange(ControlPanelFragment.STATUS_FINISH);??
???????}??
??
???????@Override??
???????public?void?onNetworkStatusChange(int?status,?Object?obj)?{??
???????????//?這里不做不論什么操作不影響簡單識別??
???????}??
???}??

獲得識別BDVRClient對象

[java]?view
plaincopy

mASREngine?=?VoiceRecognitionClient.getInstance(this);??
????????mASREngine.setTokenApis(Constants.API_KEY,?Constants.SECRET_KEY);??

請求參數(shù)設(shè)置

每次識別須要通過通過VoiceRecognitionConfig設(shè)置參數(shù)，當(dāng)中一些方法在API中有說明

開始語音識別，BDVRClient在開始識別后。會啟動錄音、預(yù)處理、上傳到server并獲取識別結(jié)果。

[java]?view
plaincopy

int?code?=?mASREngine.startVoiceRecognition(mListener,?config);??
???????????????if?(code?!=?VoiceRecognitionClient.START_WORK_RESULT_WORKING)?{??
???????????????????mResult.setText(getString(R.string.error_start,?code));??
???????????????}???

取消語音識別

[java]?view
plaincopy

mASREngine.stopVoiceRecognition();??

結(jié)束語音識別

[java]?view
plaincopy

mRecognitionClient.speakFinish();??