仿QQ语言变声功能

仿QQ语言变声功能

这次写的是QQ语言变声功能，这个功能想必大家都使用过，那么这个功能是怎么实现的呢？
在开发中一边触及和语音，视频，算法等等都和c/c++相关，我们这里也是使用的NDK，链接第三方动态库，生成我们自己的库文件。

先来讲讲第三方的动态库，我们这里使用的FMOD音频引擎提供给android的api和动态库。我们可以到官网下载：
FMOD,因为下载需要注册，但是我天朝的。。。，需要vnp,所以这边也提供了下载地址：FMOD android下的开发

想看下效果图：

我么进入正题：

1.从fmod开发包中，引入动态库和jar包,头文件

2.编写CMakeLists.txt文件

cmake_minimum_required(VERSION 3.4.1)

set(distribution_dir ../../../../libs )




find_library(   log-lib
                log )


add_library(    fmod
                SHARED
                 IMPORTED )


set_target_properties(  fmod
                        PROPERTIES IMPORTED_LOCATION
                        ${distribution_dir}/${ANDROID_ABI}/libfmod.so)

add_library(    fmodL
                SHARED
                IMPORTED )

set_target_properties(  fmodL
                        PROPERTIES IMPORTED_LOCATION
                        ${distribution_dir}/${ANDROID_ABI}/libfmodL.so)

set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -std=gnu++11")




add_library(    stonevoicer
                SHARED
                src/main/cpp/StoneVoicer.cpp )

include_directories(src/main/cpp/inc)

target_link_libraries(  stonevoicer
                        fmod
                        fmodL
                        ${log-lib} )

3.build.gradle中添加配置信息

 ndk{
        abiFilters "armeabi","x86","armeabi-v7a","arm64-v8a"
    }
sourceSets {
    main {
        jniLibs.srcDirs = ['libs']
    }
}

4.编写java代码

MainActivity:

public class MainActivity extends AppCompatActivity {

    @Override
    protected void onCreate(Bundle savedInstanceState) {
        super.onCreate(savedInstanceState);
        setContentView(R.layout.activity_main);
        requestPermission();
        FMOD.init(this);
    }
    private void  requestPermission(){

        if(ContextCompat.checkSelfPermission(this, Manifest.permission.WRITE_EXTERNAL_STORAGE)!= PackageManager.PERMISSION_GRANTED) {
            ActivityCompat.requestPermissions(this, new String[]{Manifest.permission.WRITE_EXTERNAL_STORAGE}, 100);
        }
        if(ContextCompat.checkSelfPermission(this,Manifest.permission.RECORD_AUDIO)!= PackageManager.PERMISSION_GRANTED) {
            ActivityCompat.requestPermissions(this, new String[]{Manifest.permission.RECORD_AUDIO}, 200);
        }
    }

    public void mFix(View v) {
        //音频路径可以去获取麦克风的数据，我这里写死了
        String path = "file:///android_asset/xiaoxin.wav";
        switch (v.getId()) {
            case R.id.btn_normal:
                VoicerUtil.fix(path, VoicerUtil.MODE_NORMAL);
                break;
            case R.id.btn_luoli:
                VoicerUtil.fix(path, VoicerUtil.MODE_LUOLI);
                break;
            case R.id.btn_dashu:
                VoicerUtil.fix(path, VoicerUtil.MODE_DASHU);
                break;
            case R.id.btn_jingsong:
                VoicerUtil.fix(path, VoicerUtil.MODE_JINGSONG);
                break;
            case R.id.btn_gaoguai:
                VoicerUtil.fix(path, VoicerUtil.MODE_GAOGUAI);
                break;
            case R.id.btn_kongling:
                VoicerUtil.fix(path, VoicerUtil.MODE_KONGLING);
                break;
            default:
                break;
        }
    }

    @Override
    protected void onDestroy() {
        super.onDestroy();
        FMOD.close();
    }
}

VoicerUtil:

public class VoicerUtil {


    public static final int MODE_NORMAL=0;
    public static final int MODE_LUOLI=1;
    public static final int MODE_DASHU=2;
    public static final int MODE_JINGSONG=3;
    public static final int MODE_GAOGUAI=4;
    public static final int MODE_KONGLING=5;
    /**
     *
     * @param path
     * @param type
     */
    public native static void fix(String path,int type);
    static{
        System.loadLibrary("fmod");
        System.loadLibrary("fmodL");
        System.loadLibrary("stonevoicer");
    }
}

5.生成头文件

我这边使用的是javah来生成的：

javah -classpath . -jni github.com.stonevoicer.VoicerUtil

然后将头文件移动到cpp文件夹下：

6.编写c++代码

#include <jni.h>
#include <string>
#include <stdlib.h>
#include "github_com_stonevoicer_VoicerUtil.h"
#include <fmod.hpp>
#include <unistd.h>

#include <android/log.h>
#define LOGI(FORMAT,...) __android_log_print(ANDROID_LOG_INFO,"zph",FORMAT,##__VA_ARGS__);
#define LOGE(FORMAT,...) __android_log_print(ANDROID_LOG_ERROR,"zph",FORMAT,##__VA_ARGS__);


#define MODE_NORMAL 0
#define MODE_LUOLI 1
#define MODE_DASHU 2
#define MODE_JINGSONG 3
#define MODE_GAOGUAI 4
#define MODE_KONGLING 5

using namespace FMOD;

JNIEXPORT void JNICALL Java_github_com_stonevoicer_VoicerUtil_fix
        (JNIEnv *env, jclass jcls, jstring path_jstr, jint mode){
    LOGI("%s", "===>start");
    System * system;
    Sound *sound;
    DSP *dsp;
    Channel *channel;
    float frequency;
    bool isplaying= true;
    System_Create(&system);
    system->init(32,FMOD_INIT_NORMAL,NULL);
    const char* path_cstr = env->GetStringUTFChars(path_jstr,NULL);
    try {
        system->createSound(path_cstr,FMOD_DEFAULT,NULL,&sound);
        switch (mode){
            case MODE_NORMAL:
                LOGI("%s", path_cstr);
                system->playSound(sound, 0, false, &channel);
                LOGI("%s", "fix normal");
                break;
            case MODE_LUOLI:
                //dsp->音效
                //dsp 提升或者降低音调的一种音效
                system->createDSPByType(FMOD_DSP_TYPE_PITCHSHIFT, &dsp);
                dsp->setParameterFloat(FMOD_DSP_PITCHSHIFT_PITCH,8.0);
                system->playSound(sound, 0, false, &channel);
                channel->addDSP(0, dsp);
                break;
            case MODE_DASHU:
                system->createDSPByType(FMOD_DSP_TYPE_PITCHSHIFT, &dsp);
                dsp->setParameterFloat(FMOD_DSP_PITCHSHIFT_PITCH,0.8);
                system->playSound(sound, 0, false, &channel);
                channel->addDSP(0, dsp);
                break;
            case MODE_JINGSONG:
                system->createDSPByType(FMOD_DSP_TYPE_TREMOLO, &dsp);
                dsp->setParameterFloat(FMOD_DSP_TREMOLO_SKEW, 5);
                system->playSound(sound, 0, false, &channel);
                channel->addDSP(0, dsp);
                break;
            case MODE_GAOGUAI:
                //提高说话的速度
                system->playSound(sound, 0, false, &channel);
                channel->getFrequency(&frequency);
                frequency = frequency * 2;
                channel->setFrequency(frequency);
                break;
            case MODE_KONGLING:
                //空灵
                system->createDSPByType(FMOD_DSP_TYPE_ECHO, &dsp);
                dsp->setParameterFloat(FMOD_DSP_ECHO_DELAY, 300);
                dsp->setParameterFloat(FMOD_DSP_ECHO_FEEDBACK, 20);
                system->playSound(sound, 0, false, &channel);
                channel->addDSP(0, dsp);
                break;
        }

    }catch (...){
        LOGE("%s", "发生异常");
        goto end;
    }
    system->update();

    while (isplaying){
        channel->isPlaying(&isplaying);
        usleep(1000 * 1000);
    }


    goto end;


    end:
    env->ReleaseStringUTFChars(path_jstr, path_cstr);
    sound->release();
    system->close();
    system->release();

}

下面给出项目地址：StoneVoicer-Demo