Android音视频学习第7章：使用OpenSL ES进行音频解码

Posted 2020-08-29 薛萌

/*
*
*这里使用了transcode-1.1.7对wav文件进行解码，然后使用opensl es进行播放
*
*/

---

//用到的变量和结构体

WAV wav; //wav文件指针
SLObjectItf engineObject; //引擎对象
SLEngineItf engineInterface; //引擎接口
SLObjectItf outputMixObject; //混音器
SLObjectItf audioPlayerObject; //播放器对象
SLandroidSimpleBufferQueueItf andioPlayerBufferQueueItf;    //缓冲器队列接口
SLPlayItf audioPlayInterface;   //播放接口


unsigned char *buffer; //缓冲区
size_t bufferSize;     //缓冲区大小

//上下文
struct PlayerContext{
    WAV wav;
    unsigned char *buffer;
    size_t bufferSize;

    PlayerContext(WAV wav,
            unsigned char *buffer,
            size_t bufferSize){
        this->wav = wav;
        this->buffer = buffer;
        this->bufferSize = bufferSize;
    }
};
//实现对象
void RealizeObject(SLObjectItf object){
    //非异步（阻塞）
    (*object)->Realize(object,SL_BOOLEAN_FALSE);
}

具体实现流程：

1.打开文件

WAV wav = OpenWaveFile(env,jFileName);

//打开文件
WAV OpenWaveFile(JNIEnv *env,jstring jFileName){
    const char *cFileName = env->GetStringUTFChars(jFileName,JNI_FALSE);
    WAVError err;
    WAV wav = wav_open(cFileName,WAV_READ,&err);

    LOGI("%d",wav_get_bitrate(wav));
    env->ReleaseStringUTFChars(jFileName,cFileName);
    if(wav == 0){
        LOGE("%s",wav_strerror(err));
    }
    return wav;
}

2.创建OpenSL ES引擎

//OpenSL ES在Android平台下默认是线程安全的，这样设置是为了为了兼容其他平台
    SLEngineOption options[] = {
        {(SLuint32)SL_ENGINEOPTION_THREADSAFE, (SLuint32)SL_BOOLEAN_TRUE}
    };
    slCreateEngine(&engineObject,ARRAY_LEN(engineObject),options,0,0,0); //没有接口

//实例化对象
    //对象创建之后，处于未实例化状态，对象虽然存在但未分配任何资源，使用前先实例化（使用完之后destroy）
    RealizeObject(engineObject);

3.获取引擎接口

(*engineObject)->GetInterface(engineObject,SL_IID_ENGINE,&engineInterface);

4.创建输出混音器

(*engineInterface)->CreateOutputMix(engineInterface,&outputMixObject,0,0,0); //没有接口
    //实例化混音器
    RealizeObject(outputMixObject);

5.创建缓冲区保存读取到的音频数据库

//缓冲区的大小
    bufferSize = wav_get_channels(wav) * wav_get_rate(wav) * wav_get_bits(wav);
    buffer = new unsigned char[bufferSize];

6.创建带有缓冲区队列的音频播放器

CreateBufferQueueAudioPlayer(wav,engineInterface,outputMixObject,audioPlayerObject);
    //实例化音频播放器
    RealizeObject(audioPlayerObject);

CreateBufferQueueAudioPlayer.cpp

extern "C" {
#include "wavlib.h"
}
#include <SLES/OpenSLES.h>
#include <SLES/OpenSLES_Android.h>
#include <android/log.h>

#define ARRAY_LEN(a) (sizeof(a) / sizeof(a[0]))
//创建音频播放对象
void CreateBufferQueueAudioPlayer(
        WAV wav,
        SLEngineItf engineEngine,
        SLObjectItf outputMixObject,
        SLObjectItf &audioPlayerObject){

    // Android针对数据源的简单缓冲区队列定位器
    SLDataLocator_AndroidSimpleBufferQueue dataSourceLocator = {
        SL_DATALOCATOR_ANDROIDSIMPLEBUFFERQUEUE, // 定位器类型
        1                                        // 缓冲区数
    };

    // PCM数据源格式
    SLDataFormat_PCM dataSourceFormat = {
        SL_DATAFORMAT_PCM,        // 格式类型
        wav_get_channels(wav),    // 通道数
        wav_get_rate(wav) * 1000, // 毫赫兹/秒的样本数
        wav_get_bits(wav),        // 每个样本的位数
        wav_get_bits(wav),        // 容器大小
        SL_SPEAKER_FRONT_CENTER,  // 通道屏蔽
        SL_BYTEORDER_LITTLEENDIAN // 字节顺序
    };

    // 数据源是含有PCM格式的简单缓冲区队列
    SLDataSource dataSource = {
        &dataSourceLocator, // 数据定位器
        &dataSourceFormat   // 数据格式
    };

    // 针对数据接收器的输出混合定位器
    SLDataLocator_OutputMix dataSinkLocator = {
        SL_DATALOCATOR_OUTPUTMIX, // 定位器类型
        outputMixObject           // 输出混合
    };

    // 数据定位器是一个输出混合
    SLDataSink dataSink = {
        &dataSinkLocator, // 定位器
        0                 // 格式
    };

    // 需要的接口
    SLInterfaceID interfaceIds[] = {
        SL_IID_BUFFERQUEUE
    };

    // 需要的接口，如果所需要的接口不要用，请求将失败
    SLboolean requiredInterfaces[] = {
        SL_BOOLEAN_TRUE // for SL_IID_BUFFERQUEUE
    };

    // 创建音频播放器对象
    SLresult result = (*engineEngine)->CreateAudioPlayer(
            engineEngine,
            &audioPlayerObject,
            &dataSource,
            &dataSink,
            ARRAY_LEN(interfaceIds),
            interfaceIds,
            requiredInterfaces);
}

7.获得缓冲区队列接口Buffer Queue Interface

//通过缓冲区队列接口对缓冲区进行排序播放
    (*audioPlayerObject)->GetInterface(audioPlayerObject,SL_IID_BUFFERQUEUE,&andioPlayerBufferQueueItf);

8.注册音频播放器回调函数

//当播放器完成对前一个缓冲区队列的播放时，回调函数会被调用，然后我们又继续读取音频数据，直到结束
    //上下文，包裹参数方便再回调函数中使用
    PlayerContext *ctx = new PlayerContext(wav,buffer,bufferSize);
    (*andioPlayerBufferQueueItf)->RegisterCallback(andioPlayerBufferQueueItf,PlayerCallBack,ctx);

//回调函数
void PlayerCallBack(SLAndroidSimpleBufferQueueItf andioPlayerBufferQueue,void *context){
    PlayerContext* ctx = (PlayerContext*)context;
    //读取数据
    ssize_t readSize = wav_read_data(ctx->wav,ctx->buffer,ctx->bufferSize);
    if(0 < readSize){
        (*andioPlayerBufferQueue)->Enqueue(andioPlayerBufferQueue,ctx->buffer,readSize);
    }else{
        //destroy context
        CloseWaveFile(ctx->wav); //关闭文件
        delete ctx->buffer; //释放缓存
    }
}

9.获取Play Interface通过对SetPlayState函数来启动播放音乐

//一旦播放器被设置为播放状态，该音频播放器开始等待缓冲区排队就绪
    (*audioPlayerObject)->GetInterface(audioPlayerObject,SL_IID_PLAY,&audioPlayInterface);
    //设置播放状态
    (*audioPlayInterface)->SetPlayState(audioPlayInterface,SL_PLAYSTATE_PLAYING);

10.开始，让第一个缓冲区入队

PlayerCallBack(andioPlayerBufferQueueItf,ctx);

完整代码

#include "com_dongnaoedu_jasonaudioplayer_AudioPlayer.h"

extern "C" {
#include "wavlib.h"
}
#include <SLES/OpenSLES.h>
#include <SLES/OpenSLES_Android.h>
#include <android/log.h>
#include "CreateBufferQueueAudioPlayer.cpp"

#define LOGI(FORMAT,...) __android_log_print(ANDROID_LOG_INFO,"jason",FORMAT,##__VA_ARGS__);
#define LOGE(FORMAT,...) __android_log_print(ANDROID_LOG_ERROR,"jason",FORMAT,##__VA_ARGS__);


#define ARRAY_LEN(a) (sizeof(a) / sizeof(a[0]))


WAV wav; //wav文件指针
SLObjectItf engineObject; //引擎对象
SLEngineItf engineInterface; //引擎接口
SLObjectItf outputMixObject; //混音器
SLObjectItf audioPlayerObject; //播放器对象
SLAndroidSimpleBufferQueueItf andioPlayerBufferQueueItf;    //缓冲器队列接口
SLPlayItf audioPlayInterface;   //播放接口


unsigned char *buffer; //缓冲区
size_t bufferSize;     //缓冲区大小

//上下文
struct PlayerContext{
    WAV wav;
    unsigned char *buffer;
    size_t bufferSize;

    PlayerContext(WAV wav,
            unsigned char *buffer,
            size_t bufferSize){
        this->wav = wav;
        this->buffer = buffer;
        this->bufferSize = bufferSize;
    }
};

//打开文件
WAV OpenWaveFile(JNIEnv *env,jstring jFileName){
    const char *cFileName = env->GetStringUTFChars(jFileName,JNI_FALSE);
    WAVError err;
    WAV wav = wav_open(cFileName,WAV_READ,&err);

    LOGI("%d",wav_get_bitrate(wav));
    env->ReleaseStringUTFChars(jFileName,cFileName);
    if(wav == 0){
        LOGE("%s",wav_strerror(err));
    }
    return wav;
}

//关闭文件
void CloseWaveFile(WAV wav){
    wav_close(wav);
}

//实现对象
void RealizeObject(SLObjectItf object){
    //非异步（阻塞）
    (*object)->Realize(object,SL_BOOLEAN_FALSE);
}

//回调函数
void PlayerCallBack(SLAndroidSimpleBufferQueueItf andioPlayerBufferQueue,void *context){
    PlayerContext* ctx = (PlayerContext*)context;
    //读取数据
    ssize_t readSize = wav_read_data(ctx->wav,ctx->buffer,ctx->bufferSize);
    if(0 < readSize){
        (*andioPlayerBufferQueue)->Enqueue(andioPlayerBufferQueue,ctx->buffer,readSize);
    }else{
        //destroy context
        CloseWaveFile(ctx->wav); //关闭文件
        delete ctx->buffer; //释放缓存
    }
}

JNIEXPORT void JNICALL Java_com_dongnaoedu_jasonaudioplayer_AudioPlayer_play
  (JNIEnv *env, jclass jthiz, jstring jFileName){
    //1.打开文件
    WAV wav = OpenWaveFile(env,jFileName);

    //2.创建OpenSL ES引擎
    //OpenSL ES在Android平台下默认是线程安全的，这样设置是为了为了兼容其他平台
    SLEngineOption options[] = {
        {(SLuint32)SL_ENGINEOPTION_THREADSAFE, (SLuint32)SL_BOOLEAN_TRUE}
    };
    slCreateEngine(&engineObject,ARRAY_LEN(engineObject),options,0,0,0); //没有接口

    //实例化对象
    //对象创建之后，处于未实例化状态，对象虽然存在但未分配任何资源，使用前先实例化（使用完之后destroy）
    RealizeObject(engineObject);

    //3.获取引擎接口
    (*engineObject)->GetInterface(engineObject,SL_IID_ENGINE,&engineInterface);

    //4.创建输出混音器
    (*engineInterface)->CreateOutputMix(engineInterface,&outputMixObject,0,0,0); //没有接口

    //实例化混音器
    RealizeObject(outputMixObject);

    //5.创建缓冲区保存读取到的音频数据库
    //缓冲区的大小
    bufferSize = wav_get_channels(wav) * wav_get_rate(wav) * wav_get_bits(wav);
    buffer = new unsigned char[bufferSize];

    //6.创建带有缓冲区队列的音频播放器
    CreateBufferQueueAudioPlayer(wav,engineInterface,outputMixObject,audioPlayerObject);

    //实例化音频播放器
    RealizeObject(audioPlayerObject);

    //7.获得缓冲区队列接口Buffer Queue Interface
    //通过缓冲区队列接口对缓冲区进行排序播放
    (*audioPlayerObject)->GetInterface(audioPlayerObject,SL_IID_BUFFERQUEUE,&andioPlayerBufferQueueItf);

    //8.注册音频播放器回调函数
    //当播放器完成对前一个缓冲区队列的播放时，回调函数会被调用，然后我们又继续读取音频数据，直到结束
    //上下文，包裹参数方便再回调函数中使用
    PlayerContext *ctx = new PlayerContext(wav,buffer,bufferSize);
    (*andioPlayerBufferQueueItf)->RegisterCallback(andioPlayerBufferQueueItf,PlayerCallBack,ctx);

    //9.获取Play Interface通过对SetPlayState函数来启动播放音乐
    //一旦播放器被设置为播放状态，该音频播放器开始等待缓冲区排队就绪
    (*audioPlayerObject)->GetInterface(audioPlayerObject,SL_IID_PLAY,&audioPlayInterface);
    //设置播放状态
    (*audioPlayInterface)->SetPlayState(audioPlayInterface,SL_PLAYSTATE_PLAYING);

    //10.开始，让第一个缓冲区入队
    PlayerCallBack(andioPlayerBufferQueueItf,ctx);


    //关闭文件
    //CloseWaveFile(wav);
}