當(dāng)前位置：首頁(yè) > 编程资源 > 编程问答 >内容正文

编程问答

语音识别，语义理解一站式解决（android平台olami sdk）

發(fā)布時(shí)間：2025/3/15 编程问答 25 豆豆

生活随笔收集整理的這篇文章主要介紹了语音识别，语义理解一站式解决（android平台olami sdk）小編覺(jué)得挺不錯(cuò)的,現(xiàn)在分享給大家,幫大家做個(gè)參考.

語(yǔ)音記賬demo：http://blog.csdn.net/ls0609/article/details/72765789

olami sdk實(shí)現(xiàn)了把錄音或者文字轉(zhuǎn)化為用戶可以理解的json字符串從而實(shí)現(xiàn)語(yǔ)義理解，用戶可以定義自己的語(yǔ)義，是不是很強(qiáng)大？本文講述怎么自定義語(yǔ)義，以及如何解析自定義語(yǔ)義。本文使用olami sdk做了一個(gè)在線聽(tīng)書(shū)的demo，用戶只需類似“我想聽(tīng)***”就能實(shí)現(xiàn)聽(tīng)書(shū)的在線查找并播放。用的是喜馬拉雅的在線聽(tīng)書(shū)sdk.基于eclipse開(kāi)發(fā)環(huán)境，libs目錄下jar和so文件如下：olami-android-sdk.jar //olami sdk 的jar afinal_0.5.1_bin.jar litepal.jar gson-2.2.4.jar okhttp-2.4.0.jar okhttp-urlconnection-2.2.0.jar okio-1.4.0.jar opensdk.jar //上面這幾個(gè)都是喜馬拉雅需要的jar libspeex.so //olami sdk 需要用到speex壓縮功能 libxmediaplayer.so // 喜馬拉雅so libxmediaplayer_x.so // 喜馬拉雅so

概述：?
VoiceSdkService中定義了OlamiVoiceRecognizer語(yǔ)音識(shí)別引擎，通過(guò)點(diǎn)擊MusicActivity的開(kāi)始button啟動(dòng)錄音，錄音結(jié)果在VoiceSdkService中的onResult()回調(diào)中拿到識(shí)別的Json字符串，在processServiceMessage()函數(shù)中處理后找到要聽(tīng)書(shū)的名稱，然后進(jìn)入BookUtil進(jìn)行搜索，搜索到結(jié)果后通知VoiceSdkService進(jìn)行播放，并通知MusicActivity更新播放進(jìn)度等信息。

1.AndroidManifest.xml配置

<?xml version="1.0" encoding="utf-8"?><manifest xmlns:android="http://schemas.android.com/apk/res/android"package="com.olami.musicdemo"android:versionCode="1"android:versionName="1.0" ><uses-sdk android:minSdkVersion="8"android:targetSdkVersion="21" /><uses-permission android:name="android.permission.RECORD_AUDIO"/> <uses-permission android:name="android.permission.INTERNET"/> <uses-permission android:name="android.permission.ACCESS_NETWORK_STATE"/> <uses-permission android:name="android.permission.ACCESS_WIFI_STATE"/> <uses-permission android:name="android.permission.READ_PHONE_STATE"/><uses-permission android:name="android.permission.READ_EXTERNAL_STORAGE" /><uses-permission android:name="android.permission.WRITE_EXTERNAL_STORAGE" /><application android:name="com.olami.musicdemo.OlamiApplication"android:allowBackup="true"android:icon="@drawable/ic_launcher"android:label="@string/app_name"android:theme="@style/AppTheme" ><meta-data android:name="app_key"android:value="b617866c20482d133d5de66fceb37da3" /><meta-data android:name="pack_id"android:value="com.app.test.android" /><activity android:name=".MusicActivity"android:label="@string/app_name" ><intent-filter><action android:name="android.intent.action.MAIN" /><category android:name="android.intent.category.LAUNCHER" /></intent-filter></activity><service android:name=".VoiceSdkService"android:exported="true" ></service><service android:name="com.ximalaya.ting.android.opensdk.player.service.XmPlayerService"/></application></manifest>

2.layout布局文件

layout_musicview.xml

TextView 有兩個(gè)，tv_name顯示聽(tīng)書(shū)的名稱, tv_totoal_time顯示聽(tīng)書(shū)的總時(shí)間。?
ProgressBar 實(shí)時(shí)刷新顯示聽(tīng)書(shū)的進(jìn)度

activity_music.xml

自定義MusicView比較簡(jiǎn)單，代碼如下：

public class MusicView extends RelativeLayout{private Context mContext;private Handler mHandler;private TextView mTextViewName;private TextView mTextViewTotalTime;private ProgressBar mProgressBar;public MusicView(Context context,AttributeSet attrs) {super(context,attrs);LayoutInflater inflater =(LayoutInflater) context.getSystemService(context.LAYOUT_INFLATER_SERVICE);RelativeLayout view = (RelativeLayout) inflater.inflate(R.layout.layout_musicview, this,true);mTextViewName = (TextView) view.findViewById(R.id.tv_name);mTextViewTotalTime = (TextView) view.findViewById(R.id.tv_total_time);mProgressBar = (ProgressBar)view.findViewById(R.id.progressbar_music);}public void initMusicView(Context context,Handler handler){mContext = context;mHandler = handler;}public void setMusicName(String name){//設(shè)置播放名稱mTextViewName.setText(name);}public void setProgress(int progress){//設(shè)置播放進(jìn)度mProgressBar.setProgress(progress);}public void setTotalTime(String time){//設(shè)置播放總時(shí)間mTextViewTotalTime.setText(time);}}

布局效果圖如下：?

3.MusicActivity和VoiceSdkService通信

本文沒(méi)有用bind service的方式實(shí)現(xiàn)activity和service的消息通信。?
MusicAcitity 和 VoiceSdkService中分別實(shí)現(xiàn)了一個(gè)CommunicationAssist的接口

public interface CommunicationAssist {public void callBack(int what, int arg1, int arg2,Bundle data, Object obj); }

然后把他們分別實(shí)現(xiàn)CommunicationAssist接口的變量注冊(cè)到OlamiApplication，這樣通過(guò)OlamiApplication實(shí)現(xiàn)了MusicAcitity 和 VoiceSdkService橋接。

3.1OlamiApplication

1) 注冊(cè)MusicActivity到VoiceSdkService的回調(diào)

public void setActivityToServiceListener(CommunicationAssist listener) {ActivityToServiceListener = listener; }

這個(gè)是在VoiceSdkService中調(diào)用setActivityToServiceListener(),把VoiceSdkService中的VoiceSdkComAssist注冊(cè)到application中，MusicActivity中可以通過(guò)getActivityToServiceListener?
這個(gè)函數(shù)回調(diào)向VoiceSdkService發(fā)送消息。

2) 注冊(cè) VoiceSdkService到MusicActivity回調(diào)

public void setServiceToActivityListener(CommunicationAssist listener) {mServiceToActivityListener = listener; }

這個(gè)是在MusicAcitivity中調(diào)用setServiceToActivityListener(),這樣在VoiceSdkService中就可以通過(guò)getServiceToActivityListener()獲得回調(diào)向MusciActivity發(fā)送消息。

3.2 MusicActivity

public class MusicActivity extends Activity {private Handler mHandler;private Handler mInComingHandler;private ActivityComAssist mActivityComAssist;private Button mBtnStart;private Button mBtnStop;private Button mBtnCancel;private Button mBtnSend;private EditText mEditText;private TextView mTextView;private TextView mInputTextView;private TextView mTextViewVolume;private BookUtil mBookUtil = null;private MusicView mMusicView = null;@Overrideprotected void onCreate(Bundle savedInstanceState) {super.onCreate(savedInstanceState);setContentView(R.layout.activity_music);initHandler();//初始化handler用于內(nèi)部消息處理initInComingHandler();//用于處理來(lái)自VoiceSdkService的消息initCommunicationAssist();//向application注冊(cè)消息回調(diào)，VoiceSdkSerive可以//通過(guò)getServiceToActivityListener()獲得回調(diào)向MusicActivity發(fā)送消息initView();//初始化view控件Intent intent = new Intent();intent.setClass(MusicActivity.this, VoiceSdkService.class);startService(intent);//啟動(dòng)后臺(tái)服務(wù)}private void initView(){mBtnStart = (Button) findViewById(R.id.btn_start);mBtnStop = (Button) findViewById(R.id.btn_stop);mBtnCancel = (Button) findViewById(R.id.btn_cancel);mBtnSend = (Button) findViewById(R.id.btn_send);mInputTextView = (TextView) findViewById(R.id.tv_inputText);mEditText = (EditText) findViewById(R.id.et_content);mTextView = (TextView) findViewById(R.id.tv_result);mTextViewVolume = (TextView) findViewById(R.id.tv_volume);mBtnStart.setOnClickListener(new OnClickListener(){@Overridepublic void onClick(View v) {sendMessageToService(MessageConst.CLIENT_ACTION_START_RECORED,0,0,null,null);} });mBtnStop.setOnClickListener(new OnClickListener(){@Overridepublic void onClick(View v) {sendMessageToService(MessageConst.CLIENT_ACTION_STOP_RECORED,0,0,null,null);mBtnStart.setText("開(kāi)始");Log.i("led","MusicActivity mBtnStop onclick 開(kāi)始");} });mBtnCancel.setOnClickListener(new OnClickListener(){@Overridepublic void onClick(View v) {sendMessageToService(MessageConst.CLIENT_ACTION_CANCEL_RECORED,0,0,null,null);} });mBtnSend.setOnClickListener(new OnClickListener(){@Overridepublic void onClick(View v) {sendMessageToService(MessageConst.CLIENT_ACTION_SENT_TEXT,0,0,null,mEditText.getText());mInputTextView.setText("文字: "+mEditText.getText());} });mMusicView = (MusicView) findViewById(R.id.music_view);//if(mMusicView != null)//mMusicView.initMusicView(MusicActivity.this,mHandler);}private void initHandler(){mHandler = new Handler(){@Overridepublic void handleMessage(Message msg){switch (msg.what){case MessageConst.CLIENT_ACTION_START_RECORED:break;default:break; }}};}//InComingHandler 收到來(lái)自VoiceSdkService的消息用于更新界面，//包括開(kāi)始錄音，結(jié)束錄音，播放的書(shū)的名稱和進(jìn)度，音量等信息。private void initInComingHandler(){mInComingHandler = new Handler(){@Overridepublic void handleMessage(Message msg){switch (msg.what){case MessageConst.CLIENT_ACTION_START_RECORED:mBtnStart.setText("錄音中");Log.i("led","MusicActivity 錄音中");break;case MessageConst.CLIENT_ACTION_STOP_RECORED:mBtnStart.setText("識(shí)別中");Log.i("led","MusicActivity 識(shí)別中");break;case MessageConst.CLIENT_ACTION_CANCEL_RECORED:mBtnStart.setText("開(kāi)始");mTextView.setText("已取消");break;case MessageConst.CLIENT_ACTION_ON_ERROR:mTextView.setText("錯(cuò)誤代碼："+msg.arg1);mBtnStart.setText("開(kāi)始");break;case MessageConst.CLIENT_ACTION_UPDATA_VOLUME:mTextViewVolume.setText("音量: "+msg.arg1);break;case MessageConst.SERVER_ACTION_RETURN_RESULT://mTextView.setText(msg.obj.toString());mBtnStart.setText("開(kāi)始");break;case MessageConst.CLIENT_ACTION_PLAY_BOOK_AFTER_SEARCH:mBtnStart.setText("開(kāi)始");mBookUtil = BookUtil.getInstance();mBookUtil.play(msg.arg1);break;case MessageConst.CLIENT_ACTION_UPDATA_PLAYING_BOOK_NAME:mMusicView.setMusicName(msg.obj.toString());break;case MessageConst.CLIENT_ACTION_UPDATE_BOOK_PROGRESS:int current = msg.arg1;int duration = msg.arg2;mMusicView.setProgress(current*100/duration);float time = duration/1000/60;mMusicView.setTotalTime("總時(shí)間："+time);break;case MessageConst.CLIENT_ACTION_UPDATA_INPUT_TEXT:if(msg.obj != null)mInputTextView.setText("文字: "+msg.obj.toString());break;case MessageConst.CLIENT_ACTION_UPDATA_SERVER_MESSAGE:if(msg.obj != null)mTextView.setText("服務(wù)器返回sentence: "+msg.obj.toString());break;default:break;}}};}private void initCommunicationAssist(){//向Application注冊(cè)VoiceSdkService到MusicActivity的回調(diào)mActivityComAssist = new ActivityComAssist();OlamiApplication.getInstance().setServiceToActivityListener(mActivityComAssist);}private void sendMessageToService(int what, int arg1, int arg2, Bundle data, Object obj){//向VoiceSdkService發(fā)送消息if(OlamiApplication.getInstance().getActivityToServiceListener() != null)OlamiApplication.getInstance().getActivityToServiceListener().callBack(what, arg1, arg2, data, obj);}private class ActivityComAssist implements CommunicationAssist{//實(shí)現(xiàn)CommunicationAssist借口，用于回調(diào)VoiceSdkService發(fā)送過(guò)來(lái)的消息@Overridepublic void callBack(int what, int arg1, int arg2, Bundle data,Object obj) {Message msg = Message.obtain(null, what);msg.arg1 = arg1;msg.arg2 = arg2;if (data != null)msg.setData(data);if (obj != null)msg.obj = obj;mInComingHandler.sendMessage(msg);} }@Overridepublic void onDestroy() {//退出應(yīng)用，停止VoiceSdkService，會(huì)進(jìn)行資源的釋放super.onDestroy();Intent intent = new Intent();intent.setClass(MusicActivity.this, VoiceSdkService.class);stopService(intent);} }

3.3 VoiceSdkService

@Override public void onCreate() {initHandler();//用于內(nèi)部消息處理initInComingHandler();//用于處理來(lái)自MusicActivity的消息initCommunicationAssist();//向application注冊(cè)消息回調(diào)，這樣MusicActivity可//以通過(guò)getActivityToServiceListener()回調(diào)向VoiceSdkService發(fā)送消息initViaVoiceRecognizerListener();//初始化錄音識(shí)別回調(diào)listenerinit();//olami錄音識(shí)別引擎初始化initXmly();//喜馬拉雅初始化}

public void init() {initHandler();mOlamiVoiceRecognizer = new OlamiVoiceRecognizer(VoiceSdkService.this);TelephonyManager telephonyManager=(TelephonyManager) this.getSystemService((this.getBaseContext().TELEPHONY_SERVICE);String imei=telephonyManager.getDeviceId();mOlamiVoiceRecognizer.init(imei);//設(shè)置身份標(biāo)識(shí)，可以填nullmOlamiVoiceRecognizer.setListener(mOlamiVoiceRecognizerListener);//設(shè)置識(shí)別結(jié)果回調(diào)listenermOlamiVoiceRecognizer.setLocalization(OlamiVoiceRecognizer.LANGUAGE_SIMPLIFIED_CHINESE);//設(shè)置支持的語(yǔ)音類型，優(yōu)先選擇中文簡(jiǎn)體mOlamiVoiceRecognizer.setAuthorization("51a4bb56ba954655a4fc834bfdc46af1","asr","68bff251789b426896e70e888f919a6d","nli"); //注冊(cè)Appkey，在olami官網(wǎng)注冊(cè)應(yīng)用后生成的appkey//注冊(cè)api，請(qǐng)直接填寫“asr”，標(biāo)識(shí)語(yǔ)音識(shí)別類型//注冊(cè)secret，在olami官網(wǎng)注冊(cè)應(yīng)用后生成的secret//注冊(cè)seq ，請(qǐng)?zhí)顚憽皀li”mOlamiVoiceRecognizer.setVADTailTimeout(2000);//錄音時(shí)尾音結(jié)束時(shí)間，建議填//2000ms//設(shè)置經(jīng)緯度信息，不愿上傳位置信息，可以填0 mOlamiVoiceRecognizer.setLatitudeAndLongitude(31.155364678184498,121.34882432933009); }

定義OlamiVoiceRecognizerListener

onError(int errCode)//出錯(cuò)回調(diào)，可以對(duì)比官方文檔錯(cuò)誤碼看是什么錯(cuò)誤?
onEndOfSpeech()//錄音結(jié)束?
onBeginningOfSpeech()//錄音開(kāi)始?
onResult(String result, int type)//result是識(shí)別結(jié)果JSON字符串?
onCancel()//取消識(shí)別，不會(huì)再返回識(shí)別結(jié)果?
onUpdateVolume(int volume)//錄音時(shí)的音量，1-12個(gè)級(jí)別大小音量

下面是VoiceSdkService完整代碼：

public class VoiceSdkService extends Service{private Handler mHandler;private Handler mInComingHandler;private VoiceSdkComAssist mVoiceSdkComAssist;private OlamiVoiceRecognizer mOlamiVoiceRecognizer;private OlamiVoiceRecognizerListener mOlamiVoiceRecognizerListener;private BookUtil mBookUtil = null;private boolean mIsRecordPause = false;@Overridepublic void onCreate() {initHandler();initInComingHandler();initCommunicationAssist();initViaVoiceRecognizerListener();init();initXmly();}@Overridepublic IBinder onBind(Intent intent) {// TODO Auto-generated method stubreturn null;}public void init(){initHandler();mOlamiVoiceRecognizer = new OlamiVoiceRecognizer(VoiceSdkService.this);TelephonyManager telephonyManager=(TelephonyManager) this.getSystemService(this.getBaseContext().TELEPHONY_SERVICE);String imei=telephonyManager.getDeviceId();mOlamiVoiceRecognizer.init(imei);//set null if you do not want to notify olami server.mOlamiVoiceRecognizer.setListener(mOlamiVoiceRecognizerListener);mOlamiVoiceRecognizer.setLocalization(OlamiVoiceRecognizer.LANGUAGE_SIMPLIFIED_CHINESE);mOlamiVoiceRecognizer.setAuthorization("51a4bb56ba954655a4fc834bfdc46af1","asr","68bff251789b426896e70e888f919a6d","nli"); mOlamiVoiceRecognizer.setVADTailTimeout(2000);mOlamiVoiceRecognizer.setLatitudeAndLongitude(31.155364678184498,121.34882432933009); }private void initHandler(){mHandler = new Handler(){@Overridepublic void handleMessage(Message msg){switch (msg.what){case MessageConst.CLIENT_ACTION_START_RECORED:sendMessageToActivity(MessageConst.CLIENT_ACTION_START_RECORED,0,0,null,null);break;case MessageConst.CLIENT_ACTION_STOP_RECORED:sendMessageToActivity(MessageConst.CLIENT_ACTION_STOP_RECORED,0,0,null,null);break;case MessageConst.CLIENT_ACTION_ON_ERROR:sendMessageToActivity(MessageConst.CLIENT_ACTION_ON_ERROR,msg.arg1,0,null,null);break;case MessageConst.CLIENT_ACTION_PLAY_BOOK_AFTER_SEARCH:sendMessageToActivity(MessageConst.CLIENT_ACTION_PLAY_BOOK_AFTER_SEARCH, msg.arg1, 0, null, msg.obj);break;case MessageConst.CLIENT_ACTION_UPDATA_PLAYING_BOOK_NAME:sendMessageToActivity(MessageConst.CLIENT_ACTION_UPDATA_PLAYING_BOOK_NAME, msg.arg1, 0, null, msg.obj);break;case MessageConst.CLIENT_ACTION_UPDATE_BOOK_PROGRESS:sendMessageToActivity(MessageConst.CLIENT_ACTION_UPDATE_BOOK_PROGRESS, msg.arg1, msg.arg2, null, null);break;case MessageConst.CLIENT_ACTION_CANCEL_RECORED:sendMessageToActivity(MessageConst.CLIENT_ACTION_CANCEL_RECORED, msg.arg1, msg.arg2, null, null);break;default:break;}}};}private void initInComingHandler(){mInComingHandler = new Handler(){@Overridepublic void handleMessage(Message msg){switch (msg.what){case MessageConst.CLIENT_ACTION_START_RECORED:if(mOlamiVoiceRecognizer != null)mOlamiVoiceRecognizer.start(); break;case MessageConst.CLIENT_ACTION_STOP_RECORED:if(mOlamiVoiceRecognizer != null)mOlamiVoiceRecognizer.stop(); break;case MessageConst.CLIENT_ACTION_CANCEL_RECORED:if(mOlamiVoiceRecognizer != null)mOlamiVoiceRecognizer.cancel(); break;case MessageConst.CLIENT_ACTION_SENT_TEXT:if(mOlamiVoiceRecognizer != null)mOlamiVoiceRecognizer.sendText(msg.obj.toString()); break;}}};}private void initViaVoiceRecognizerListener(){mOlamiVoiceRecognizerListener = new OlamiVoiceRecognizerListener();}private class OlamiVoiceRecognizerListener implements IOlamiVoiceRecognizerListener{@Overridepublic void onError(int errCode) {mHandler.sendMessage(mHandler.obtainMessage(MessageConst.CLIENT_ACTION_ON_ERROR,errCode,0));}@Overridepublic void onEndOfSpeech() {mHandler.sendEmptyMessage(MessageConst.CLIENT_ACTION_STOP_RECORED);if(mIsRecordPause){mIsRecordPause = false;mBookUtil.resumePlay();}}@Overridepublic void onBeginningOfSpeech() {if(mBookUtil.isPlaying()){mBookUtil.pause();mIsRecordPause = true;}mHandler.sendEmptyMessage(MessageConst.CLIENT_ACTION_START_RECORED);}@Overridepublic void onResult(String result, int type) { sendMessageToActivity(MessageConst.SERVER_ACTION_RETURN_RESULT,type,0,null,result);processServiceMessage(result);}@Overridepublic void onCancel() {mHandler.sendEmptyMessage(MessageConst.CLIENT_ACTION_CANCEL_RECORED);}@Overridepublic void onUpdateVolume(int volume) {sendMessageToActivity(MessageConst.CLIENT_ACTION_UPDATA_VOLUME,volume,0,null,null);}}private void initCommunicationAssist(){mVoiceSdkComAssist = new VoiceSdkComAssist();OlamiApplication.getInstance().setActivityToServiceListener(mVoiceSdkComAssist);}private void initXmly(){if(mBookUtil == null){mBookUtil = BookUtil.getInstance();mBookUtil.init(VoiceSdkService.this);mBookUtil.setHandler(mHandler);}}private void processServiceMessage(String message){String input = null;String serverMessage = null;try{JSONObject jsonObject = new JSONObject(message);JSONArray jArrayNli = jsonObject.optJSONObject("data").optJSONArray("nli");JSONObject jObj = jArrayNli.optJSONObject(0);JSONArray jArraySemantic = null;if(message.contains("semantic"))jArraySemantic = jObj.getJSONArray("semantic");else{input = jsonObject.optJSONObject("data").optJSONObject("asr").optString("result");sendMessageToActivity(MessageConst.CLIENT_ACTION_UPDATA_INPUT_TEXT, 0, 0, null, input);serverMessage = jObj.optJSONObject("desc_obj").opt("result").toString();sendMessageToActivity(MessageConst.CLIENT_ACTION_UPDATA_SERVER_MESSAGE, 0, 0, null, serverMessage);return;}JSONObject jObjSemantic;JSONArray jArraySlots;JSONArray jArrayModifier;String type = null;String songName = null;String singer = null;if(jObj != null) {type = jObj.optString("type");if("musiccontrol".equals(type)){jObjSemantic = jArraySemantic.optJSONObject(0);input = jObjSemantic.optString("input");jArraySlots = jObjSemantic.optJSONArray("slots");jArrayModifier = jObjSemantic.optJSONArray("modifier");String modifier = (String)jArrayModifier.opt(0);if((jArrayModifier != null) && ("play".equals(modifier))){if(jArraySlots != null)for(int i=0,k=jArraySlots.length(); i<k; i++){JSONObject obj = jArraySlots.getJSONObject(i);String name = obj.optString("name");if("singer".equals(name))singer = obj.optString("value");else if("songname".equals(name))songName = obj.optString("value");}}else if((modifier != null) && ("stop".equals(modifier))){if(mBookUtil != null)if(mBookUtil.isPlaying())mBookUtil.stop();}else if((modifier != null) && ("pause".equals(modifier))){if(mBookUtil != null)if(mBookUtil.isPlaying())mBookUtil.pause();}else if((modifier != null) && ("resume_play".equals(modifier))){if(mBookUtil != null)mBookUtil.resumePlay();}else if((modifier != null) && ("add_volume".equals(modifier))){if(mBookUtil != null)mBookUtil.addVolume();}else if((modifier != null) && ("del_volume".equals(modifier))){if(mBookUtil != null)mBookUtil.delVolume();}else if((modifier != null) && ("next".equals(modifier))){if(mBookUtil != null)mBookUtil.next();}else if((modifier != null) && ("previous".equals(modifier))){if(mBookUtil != null)mBookUtil.prev();}else if((modifier != null) && ("play_index".equals(modifier))){int position = 0;if(jArraySlots != null)for(int i=0,k=jArraySlots.length(); i<k; i++){JSONObject obj = jArraySlots.getJSONObject(i);JSONObject jNumDetial = obj.getJSONObject("num_detail");String index = jNumDetial.optString("recommend_value");position = Integer.parseInt(index) - 1;}if(mBookUtil != null)mBookUtil.skipTo(position);}}}if(songName != null){if(singer != null){}else{mBookUtil.searchBookAndPlay(songName,0,0);}}else if(singer != null){mBookUtil.searchBookAndPlay(songName,0,0);}serverMessage = jObj.optJSONObject("desc_obj").opt("result").toString();}catch (Exception e){e.printStackTrace();}//發(fā)送消息更新語(yǔ)音識(shí)別的文字sendMessageToActivity(MessageConst.CLIENT_ACTION_UPDATA_INPUT_TEXT, 0, 0, null, input);//發(fā)送消息更新服務(wù)器返回的結(jié)果字符串sendMessageToActivity(MessageConst.CLIENT_ACTION_UPDATA_SERVER_MESSAGE, 0, 0, null, serverMessage);}private void sendMessageToActivity(int what, int arg1, int arg2, Bundle data, Object obj){if(OlamiApplication.getInstance().getServiceToActivityListener() != null)OlamiApplication.getInstance().getServiceToActivityListener().callBack(what, arg1, arg2, data, obj);}private class VoiceSdkComAssist implements CommunicationAssist{@Overridepublic void callBack(int what, int arg1, int arg2, Bundle data,Object obj) {Message msg = Message.obtain(null, what);msg.arg1 = arg1;msg.arg2 = arg2;if (data != null)msg.setData(data);if (obj != null)msg.obj = obj;mInComingHandler.sendMessage(msg);} }@Overridepublic void onDestroy(){super.onDestroy();if(mOlamiVoiceRecognizer != null)mOlamiVoiceRecognizer.destroy();if(mBookUtil != null){mBookUtil.destroy();}}}

3.4 VoiceSdkService中onResult的回調(diào)處理

在VoiceSdkService.java中processServiceMessage(String message)用于處理onResult的回調(diào)數(shù)據(jù)。例如“我要聽(tīng)三國(guó)演義”返回如下數(shù)據(jù)： {"data": {"asr": {"result": "我要聽(tīng)三國(guó)演義","speech_status": 0,"final": true,"status": 0},"nli": [{"desc_obj": {"result": "正在努力搜索中，請(qǐng)稍等","status": 0},"semantic": [{"app": "musiccontrol","input": "我要聽(tīng)三國(guó)演義","slots": [{"name": "songname","value": "三國(guó)演義" }],"modifier": ["play"],"customer": "58df512384ae11f0bb7b487e"}],"type": "musiccontrol"}]},"status": "ok" }

1）解析出nli中type類型是musiccontrol,這是語(yǔ)法返回app的類型，而這個(gè)在線聽(tīng)書(shū)的demo只關(guān)心musiccontrol這個(gè)app類型，其他的忽略。

2）用戶說(shuō)的話轉(zhuǎn)成文字是在asr中的result中獲取?
3）在nli中的semantic中，input值是用戶說(shuō)的話，同asr中的result。?
modifier代表返回的行為動(dòng)作，此處可以看到是play就是要求播放，slots中的數(shù)據(jù)表示歌曲名稱是三國(guó)演義。?
那么動(dòng)作是play，內(nèi)容是歌曲名稱是三國(guó)演義，在這個(gè)demo中調(diào)用?
mBookUtil.searchBookAndPlay(songName,0,0);會(huì)先查詢，查詢到結(jié)果會(huì)再發(fā)播放消息要求播放，我要聽(tīng)三國(guó)演義這個(gè)流程就走完了。

4.BookUtil

說(shuō)一下搜索聽(tīng)書(shū)的實(shí)現(xiàn)過(guò)程 public void searchBookInfo(String bookName,final int index,final boolean isNeedPlay) {mBookName = bookName;Map<String, String> param = new HashMap<String, String>();param.put(DTransferConstants.SEARCH_KEY, bookName);param.put(DTransferConstants.CATEGORY_ID, "" + 3);//此處3代表搜索的是聽(tīng)書(shū)//param.put(DTransferConstants.PAGE, "" + mPageId);param.put(DTransferConstants.SORT, "asc");//返回列表的排序是正序還是逆序param.put(DTransferConstants.PAGE_SIZE, "" + PAGE_SIZE);//每頁(yè)能返回多少個(gè)查詢結(jié)果mPage = (index/PAGE_SIZE)+1;//當(dāng)前在第幾頁(yè)mPlayerManager = XmPlayerManager.getInstance(mContext);//喜馬拉雅初始化部分mPlayerManager.init(mNotificationId, null);mPlayerManager.addPlayerStatusListener(mPlayerStatusListener);mPlayerManager.addAdsStatusListener(mAdsListener);CommonRequest.getSearchedAlbums(param, new IDataCallBack<SearchAlbumList>(){@Overridepublic void onSuccess(SearchAlbumList object) { if (object != null && object.getAlbums() != null&& object.getAlbums().size() != 0){if (mSearchAlbumList == null){mSearchAlbumList = object;}else{mSearchAlbumList.getAlbums().addAll(object.getAlbums());}//mTrackAdapter.notifyDataSetChanged();Map<String, String> map = new HashMap<String, String>();map.put(DTransferConstants.ALBUM_ID, ""+object.getAlbums().get(0).getId());map.put(DTransferConstants.SORT, "asc");map.put(DTransferConstants.PAGE, "" + mPage);map.put(DTransferConstants.PAGE_SIZE, "" + PAGE_SIZE);CommonRequest.getTracks(map, new IDataCallBack<TrackList>(){@Overridepublic void onSuccess(TrackList object){mTrackList = object;mTotalCount = mTrackList.getTotalCount();if(mTrackList.getTracks().size() <= 0)return;String str = "專輯："+mTrackList.getAlbumTitle()+get(0).getTrackTitle().toString();if(isNeedPlay){mPosition = index % PAGE_SIZE;mHandler.sendMessage(mHandler.obtainMessage(MessageConst.CLIENT_ACTION_PLAY_BOOK_AFTER_SEARCH, index % PAGE_SIZE,0));//此處mTrackList中已經(jīng)查詢出結(jié)果//向VoiceSdkService發(fā)送消息進(jìn)行播放}elsesendBookInfoToServer(); }@Overridepublic void onError(int code, String message){Log.i("ppp","error: "+message);sendBookInfoToServer();}});}}@Overridepublic void onError(int code, String message){Log.i("ppp","error: "+message);sendBookInfoToServer();}}); }

5.demo中支持的說(shuō)法

我想聽(tīng)西游記?
我要聽(tīng)西游記?
播放西游記?
聽(tīng)西游記?
我想聽(tīng)西游記這本書(shū)?
上一首?
上一回?
下一首?
下一回?
暫停/暫停播放?
繼續(xù)/繼續(xù)播放?
聲音大一點(diǎn)?
聲音小一點(diǎn)?
關(guān)閉/關(guān)閉播放

用的是喜馬拉雅測(cè)試賬號(hào)，只支持聽(tīng)書(shū)的功能，查找歌曲的結(jié)果返回為空。

6.源碼下載鏈接

用olamisdk語(yǔ)音識(shí)別引擎做在線聽(tīng)書(shū)demo

7.相關(guān)鏈接

語(yǔ)音記賬demo：http://blog.csdn.net/ls0609/article/details/72765789

olami開(kāi)放平臺(tái)語(yǔ)法編寫簡(jiǎn)介：http://blog.csdn.net/ls0609/article/details/71624340

olami開(kāi)放平臺(tái)語(yǔ)法官方介紹：https://cn.olami.ai/wiki/?mp=nli&content=nli2.html

轉(zhuǎn)載請(qǐng)注明CSDN博文地址：http://blog.csdn.net/ls0609/article/details/71519203

總結(jié)

以上是生活随笔為你收集整理的语音识别，语义理解一站式解决（android平台olami sdk）的全部?jī)?nèi)容，希望文章能夠幫你解決所遇到的問(wèn)題。

如果覺(jué)得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。

上一篇：用Kotlin开发android平台语音
下一篇： Hinton's Dark Knowle