源码

GitHub

在线语法识别

SDK下载

灵云SDK下载

SDK集成

下载SDK以后，将jar和so导入工程

权限

<uses-permission android:name="android.permission.INTERNET" />
<uses-permission android:name="android.permission.WRITE_EXTERNAL_STORAGE" />
<uses-permission android:name="android.permission.ACCESS_NETWORK_STATE" />
<uses-permission android:name="android.permission.SYSTEM_ALERT_WINDOW" />
<uses-permission android:name="android.permission.ACCESS_COARSE_LOCATION" />
<uses-permission android:name="android.permission.ACCESS_FINE_LOCATION" />
<uses-permission android:name="android.permission.RECORD_AUDIO" />

封装

灵云配置类

package kong.qingwei.kqwhcidemo;

/**
 * Created by kqw on 2016/8/12.
 * 灵云配置信息
 */
public final class ConfigUtil {

    /**
     * 灵云APP_KEY
     */
    public static final String APP_KEY = "3d5d5466";

    /**
     * 开发者密钥
     */
    public static final String DEVELOPER_KEY = "eca643ff7b3c758745d7cf516e808d34";

    /**
     * 灵云云服务的接口地址
     */
    public static final String CLOUD_URL = "test.api.hcicloud.com:8888";

    /**
     * 需要运行的灵云能力
     */
    // 离线语音合成
    public static final String CAP_KEY_TTS_LOCAL = "tts.local.synth";
    // 云端语音合成
    public static final String CAP_KEY_TTS_CLOUD = "tts.cloud.wangjing";

    // 云端语义识别
    public static final String CAP_KEY_NUL_CLOUD = "nlu.cloud";

    // 云端自由说
    public static final String CAP_KEY_ASR_CLOUD_FREETALK = "asr.cloud.freetalk";
    // 离线自由说
    public static final String CAP_KEY_ASR_LOCAL_FREETALK = "asr.local.freetalk";
    // 云端语音识别+语义
    public static final String CAP_KEY_ASR_CLOUD_DIALOG = "asr.cloud.dialog";
    // 离线命令词
    public static final String CAP_KEY_ASR_LOCAL_GRAMMAR = "asr.local.grammar.v4";
    // 在线命令词
    public static final String CAP_KEY_ASR_CLOUD_GRAMMAR = "asr.cloud.grammar";
}

初始化灵云语音能力的工具类

package kong.qingwei.kqwhcidemo;

import android.app.Activity;
import android.os.Environment;
import android.util.Log;
import android.widget.Toast;

import com.sinovoice.hcicloudsdk.api.HciCloudSys;
import com.sinovoice.hcicloudsdk.common.AuthExpireTime;
import com.sinovoice.hcicloudsdk.common.HciErrorCode;
import com.sinovoice.hcicloudsdk.common.InitParam;

import java.io.File;
import java.text.SimpleDateFormat;
import java.util.Date;
import java.util.Locale;

/**
 * Created by kqw on 2016/8/12.
 * 初始化灵云语音
 */
public class HciUtil {

    private static final String TAG = "HciUtil";
    private Activity mActivity;
    private final String mConfigStr;

    public HciUtil(Activity activity) {
        mActivity = activity;

        // 加载信息,返回InitParam, 获得配置参数的字符串
        InitParam initParam = getInitParam();
        mConfigStr = initParam.getStringConfig();
    }

    public boolean initHci() {
        // 初始化
        int errCode = HciCloudSys.hciInit(mConfigStr, mActivity);
        if (errCode != HciErrorCode.HCI_ERR_NONE && errCode != HciErrorCode.HCI_ERR_SYS_ALREADY_INIT) {
            Toast.makeText(mActivity, "hciInit error: " + HciCloudSys.hciGetErrorInfo(errCode), Toast.LENGTH_SHORT).show();
            return false;
        }

        // 获取授权/更新授权文件 :
        errCode = checkAuthAndUpdateAuth();
        if (errCode != HciErrorCode.HCI_ERR_NONE) {
            // 由于系统已经初始化成功,在结束前需要调用方法hciRelease()进行系统的反初始化
            Toast.makeText(mActivity, "CheckAuthAndUpdateAuth error: " + HciCloudSys.hciGetErrorInfo(errCode), Toast.LENGTH_SHORT).show();
            HciCloudSys.hciRelease();
            return false;
        }
        return true;
    }

    /**
     * 释放
     */
    public void hciRelease(){
        HciCloudSys.hciRelease();
    }

    /**
     * 加载初始化信息
     *
     * @return 系统初始化参数
     */
    private InitParam getInitParam() {
        String authDirPath = mActivity.getFilesDir().getAbsolutePath();
        // 前置条件：无
        InitParam initparam = new InitParam();
        // 授权文件所在路径，此项必填
        initparam.addParam(InitParam.AuthParam.PARAM_KEY_AUTH_PATH, authDirPath);
        // 是否自动访问云授权,详见 获取授权/更新授权文件处注释
        initparam.addParam(InitParam.AuthParam.PARAM_KEY_AUTO_CLOUD_AUTH, "no");
        // 灵云云服务的接口地址，此项必填
        initparam.addParam(InitParam.AuthParam.PARAM_KEY_CLOUD_URL, ConfigUtil.CLOUD_URL);
        // 开发者Key，此项必填，由捷通华声提供
        initparam.addParam(InitParam.AuthParam.PARAM_KEY_DEVELOPER_KEY, ConfigUtil.DEVELOPER_KEY);
        // 应用Key，此项必填，由捷通华声提供
        initparam.addParam(InitParam.AuthParam.PARAM_KEY_APP_KEY, ConfigUtil.APP_KEY);
        // 配置日志参数
        String sdcardState = Environment.getExternalStorageState();
        if (Environment.MEDIA_MOUNTED.equals(sdcardState)) {
            String sdPath = Environment.getExternalStorageDirectory().getAbsolutePath();
            String packageName = mActivity.getPackageName();
            String logPath = sdPath + File.separator + "sinovoice" + File.separator + packageName + File.separator + "log" + File.separator;
            // 日志文件地址
            File fileDir = new File(logPath);
            if (!fileDir.exists()) {
                fileDir.mkdirs();
            }
            // 日志的路径，可选，如果不传或者为空则不生成日志
            initparam.addParam(InitParam.LogParam.PARAM_KEY_LOG_FILE_PATH, logPath);
            // 日志数目，默认保留多少个日志文件，超过则覆盖最旧的日志
            initparam.addParam(InitParam.LogParam.PARAM_KEY_LOG_FILE_COUNT, "5");
            // 日志大小，默认一个日志文件写多大，单位为K
            initparam.addParam(InitParam.LogParam.PARAM_KEY_LOG_FILE_SIZE, "1024");
            // 日志等级，0=无，1=错误，2=警告，3=信息，4=细节，5=调试，SDK将输出小于等于logLevel的日志信息
            initparam.addParam(InitParam.LogParam.PARAM_KEY_LOG_LEVEL, "5");
        }
        return initparam;
    }

    /**
     * 获取授权
     *
     * @return 授权结果
     */
    private int checkAuthAndUpdateAuth() {
        // 获取系统授权到期时间
        int initResult;
        AuthExpireTime objExpireTime = new AuthExpireTime();
        initResult = HciCloudSys.hciGetAuthExpireTime(objExpireTime);
        if (initResult == HciErrorCode.HCI_ERR_NONE) {
            // 显示授权日期,如用户不需要关注该值,此处代码可忽略
            Date date = new Date(objExpireTime.getExpireTime() * 1000);
            SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd", Locale.CHINA);
            Log.i(TAG, "expire time: " + sdf.format(date));
            if (objExpireTime.getExpireTime() * 1000 > System.currentTimeMillis()) {
                // 已经成功获取了授权,并且距离授权到期有充足的时间(>7天)
                Log.i(TAG, "checkAuth success");
                return initResult;
            }
        }
        // 获取过期时间失败或者已经过期
        initResult = HciCloudSys.hciCheckAuth();
        if (initResult == HciErrorCode.HCI_ERR_NONE) {
            Log.i(TAG, "checkAuth success");
            return initResult;
        } else {
            Log.e(TAG, "checkAuth failed: " + initResult);
            return initResult;
        }
    }
}

语法识别的类

和之前的语音识别一样，只不过配置了语法

在原来的基础上添加了initGrammar方法

package kong.qingwei.kqwhcidemo;

import android.app.Activity;
import android.util.Log;

import com.sinovoice.hcicloudsdk.android.asr.recorder.ASRRecorder;
import com.sinovoice.hcicloudsdk.common.asr.AsrConfig;
import com.sinovoice.hcicloudsdk.common.asr.AsrGrammarId;
import com.sinovoice.hcicloudsdk.common.asr.AsrInitParam;
import com.sinovoice.hcicloudsdk.common.asr.AsrRecogResult;
import com.sinovoice.hcicloudsdk.recorder.ASRCommonRecorder;
import com.sinovoice.hcicloudsdk.recorder.ASRRecorderListener;
import com.sinovoice.hcicloudsdk.recorder.RecorderEvent;

import java.io.IOException;
import java.io.InputStream;


/**
 * Created by kqw on 2016/8/15.
 * 语音识别类
 */
public class AsrUtil {

    private static final String TAG = "AsrUtil";
    private Activity mActivity;
    private ASRRecorder mAsrRecorder;
    private AsrConfig asrConfig;
    private OnAsrRecogListener mOnAsrRecogListener;
    private String mGrammar = null;
    private String mCapKey = ConfigUtil.CAP_KEY_ASR_CLOUD_GRAMMAR;

    public AsrUtil(Activity activity) {
        mActivity = activity;
        initAsr();
        initGrammar(mCapKey);
    }

    /**
     * 初始化语音识别
     */
    private void initAsr() {
        Log.i(TAG, "initAsr: ");
        // 初始化录音机
        mAsrRecorder = new ASRRecorder();

        // 配置初始化参数
        AsrInitParam asrInitParam = new AsrInitParam();
        String dataPath = mActivity.getFilesDir().getPath().replace("files", "lib");
        asrInitParam.addParam(AsrInitParam.PARAM_KEY_INIT_CAP_KEYS, mCapKey);
        asrInitParam.addParam(AsrInitParam.PARAM_KEY_DATA_PATH, dataPath);
        asrInitParam.addParam(AsrInitParam.PARAM_KEY_FILE_FLAG, AsrInitParam.VALUE_OF_PARAM_FILE_FLAG_ANDROID_SO);
        Log.v(TAG, "init parameters:" + asrInitParam.getStringConfig());

        // 设置初始化参数
        mAsrRecorder.init(asrInitParam.getStringConfig(), new ASRResultProcess());

        // 配置识别参数
        asrConfig = new AsrConfig();
        // PARAM_KEY_CAP_KEY 设置使用的能力
        asrConfig.addParam(AsrConfig.SessionConfig.PARAM_KEY_CAP_KEY, mCapKey);
        // PARAM_KEY_AUDIO_FORMAT 音频格式根据不同的能力使用不用的音频格式
        asrConfig.addParam(AsrConfig.AudioConfig.PARAM_KEY_AUDIO_FORMAT, AsrConfig.AudioConfig.VALUE_OF_PARAM_AUDIO_FORMAT_PCM_16K16BIT);
        // PARAM_KEY_ENCODE 音频编码压缩格式，使用OPUS可以有效减小数据流量
        asrConfig.addParam(AsrConfig.AudioConfig.PARAM_KEY_ENCODE, AsrConfig.AudioConfig.VALUE_OF_PARAM_ENCODE_SPEEX);
        // 其他配置，此处可以全部选取缺省值

        asrConfig.addParam("intention", "weather");
    }

    /**
     * 初始化语法
     *
     * @param capKey CapKey
     */
    public void initGrammar(String capKey) {
        // 语法相关的配置,若使用自由说能力可以不必配置该项
        if (capKey.contains("local.grammar")) {
            mGrammar = loadGrammar("stock_10001.gram");
            // 加载本地语法获取语法ID
            AsrGrammarId id = new AsrGrammarId();
            ASRCommonRecorder.loadGrammar("capkey=" + capKey + ",grammarType=jsgf", mGrammar, id);
            Log.d(TAG, "grammarid=" + id);
            // PARAM_KEY_GRAMMAR_TYPE 语法类型，使用自由说能力时，忽略以下此参数
            asrConfig.addParam(AsrConfig.GrammarConfig.PARAM_KEY_GRAMMAR_TYPE, AsrConfig.GrammarConfig.VALUE_OF_PARAM_GRAMMAR_TYPE_ID);
            asrConfig.addParam(AsrConfig.GrammarConfig.PARAM_KEY_GRAMMAR_ID, "" + id.getGrammarId());
        } else if (capKey.contains("cloud.grammar")) {
            mGrammar = loadGrammar("stock_10001.gram");
            // PARAM_KEY_GRAMMAR_TYPE 语法类型，使用自由说能力时，忽略以下此参数
            asrConfig.addParam(AsrConfig.GrammarConfig.PARAM_KEY_GRAMMAR_TYPE, AsrConfig.GrammarConfig.VALUE_OF_PARAM_GRAMMAR_TYPE_JSGF);
        }
    }

    /**
     * 开始语音识别
     */
    public void start(OnAsrRecogListener listener) {
        mOnAsrRecogListener = listener;
        if (mAsrRecorder.getRecorderState() == ASRRecorder.RECORDER_STATE_IDLE) {
            asrConfig.addParam(AsrConfig.SessionConfig.PARAM_KEY_REALTIME, "no");
            mAsrRecorder.start(asrConfig.getStringConfig(), mGrammar);
        } else {
            Log.i(TAG, "start: 录音机未处于空闲状态，请稍等");
        }
    }

    private class ASRResultProcess implements ASRRecorderListener {
        @Override
        public void onRecorderEventError(RecorderEvent event, int errorCode) {
            Log.i(TAG, "onRecorderEventError: errorCode = " + errorCode);
            if (null != mOnAsrRecogListener) {
                mOnAsrRecogListener.onError(errorCode);
            }
        }

        @Override
        public void onRecorderEventRecogFinsh(RecorderEvent recorderEvent, final AsrRecogResult arg1) {
            if (recorderEvent == RecorderEvent.RECORDER_EVENT_RECOGNIZE_COMPLETE) {
                Log.i(TAG, "onRecorderEventRecogFinsh: 识别结束");
            }
            if (null != mOnAsrRecogListener) {
                mActivity.runOnUiThread(new Runnable() {
                    @Override
                    public void run() {
                        mOnAsrRecogListener.onAsrRecogResult(arg1);
                    }
                });
            }
        }

        @Override
        public void onRecorderEventStateChange(RecorderEvent recorderEvent) {
            if (recorderEvent == RecorderEvent.RECORDER_EVENT_BEGIN_RECORD) {
                Log.i(TAG, "onRecorderEventStateChange: 开始录音");
            } else if (recorderEvent == RecorderEvent.RECORDER_EVENT_BEGIN_RECOGNIZE) {
                Log.i(TAG, "onRecorderEventStateChange: 开始识别");
            } else if (recorderEvent == RecorderEvent.RECORDER_EVENT_NO_VOICE_INPUT) {
                Log.i(TAG, "onRecorderEventStateChange: 无音频输入");
            } else {
                Log.i(TAG, "onRecorderEventStateChange: recorderEvent = " + recorderEvent);
            }
        }

        @Override
        public void onRecorderRecording(byte[] volumedata, int volume) {
            if (null != mOnAsrRecogListener) {
                mOnAsrRecogListener.onVolume(volume);
            }
        }

        @Override
        public void onRecorderEventRecogProcess(RecorderEvent recorderEvent, AsrRecogResult arg1) {
            if (recorderEvent == RecorderEvent.RECORDER_EVENT_RECOGNIZE_PROCESS) {
                Log.i(TAG, "onRecorderEventRecogProcess: 识别中间反馈");
            }
            if (arg1 != null) {
                if (arg1.getRecogItemList().size() > 0) {
                    Log.i(TAG, "onRecorderEventRecogProcess: 识别中间结果结果为：" + arg1.getRecogItemList().get(0).getRecogResult());
                } else {
                    Log.i(TAG, "onRecorderEventRecogProcess: 未能正确识别,请重新输入");
                }
            }
        }
    }

    /**
     * 读取语法
     *
     * @param fileName 文件名
     * @return 语法
     */
    private String loadGrammar(String fileName) {
        String grammar = "";
        InputStream is = null;
        try {
            is = mActivity.getAssets().open(fileName);
            byte[] data = new byte[is.available()];
            is.read(data);
            grammar = new String(data);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            if (is != null) {
                try {
                    is.close();
                } catch (IOException e) {
                    e.printStackTrace();
                }
            }
        }
        return grammar;
    }

    /**
     * 语音识别的回调接口
     */
    public interface OnAsrRecogListener {
        // 识别结果
        void onAsrRecogResult(AsrRecogResult asrRecogResult);

        // 识别错误码
        void onError(int errorCode);

        // 录音音量
        void onVolume(int volume);
    }
}

使用

使用和语音识别完全一致

初始化灵云的语音能力和语法识别

// 灵云语音工具类
mInitTts = new HciUtil(this);
// 初始化灵云语音
boolean isInitHci = mInitTts.initHci();
if (isInitHci) { // 初始化成功
    ……
    // 语音识别
    mAsrUtil = new AsrUtil(this);
}

语法识别

/**
 * 语音识别（语音转文字）
 *
 * @param view view
 */
public void asr(View view) {
    mAsrUtil.start(new AsrUtil.OnAsrRecogListener() {
        @Override
        public void onAsrRecogResult(AsrRecogResult asrRecogResult) {
            StringBuilder stringBuffer = new StringBuilder();
            ArrayList<AsrRecogItem> asrRecogItemArrayList = asrRecogResult.getRecogItemList();
            for (AsrRecogItem asrRecogItem : asrRecogItemArrayList) {
                String result = asrRecogItem.getRecogResult();
                Log.i(TAG, "onAsrRecogResult: " + result);
                stringBuffer.append(result).append("\n");
            }
            showDialog("识别结果", stringBuffer.toString());
        }

        @Override
        public void onError(int errorCode) {
            Log.i(TAG, "onError: " + errorCode);
        }

        @Override
        public void onVolume(int volume) {
            Log.i(TAG, "onVolume: " + volume);
        }
    });
}

离线语音识别

在在线语法识别的基础上，离线的语法识别只需要将CapKey替换，并且添加离线资源即可

离线资源的下载

下载完解压离线资源

将里面所有的文件重命名，前面加lib，后面加.so，然后导入工程

修改CapKey为asr.local.grammar.v4

注意，灵云的离线语音功能第一次使用需要联网激活，激活以后才可以使用离线功能。