Android OpenSL ES 音频采集与播放

本篇详细介绍一下基于 Android OpenSL ES 实现音频的采集与播放。

1. OpenSL ES 是什么？

OpenSL ES 是一个 针对嵌入式系统的开放硬件音频加速库 ，也可以将其视为一套针对嵌入式平台的音频标准，全称为： Open Sound Library for Embedded Systems ，它提供了一套高性能、 低延迟的音频功能实现方法，并且实现了软硬件音频性能的跨平台部署，大大降低了上层处理音频应用的开发难度。

在 Android 开发中，Google 官方从 Android 2.3 （API 9）开始，便支持了 OpenSL ES 标准 ，并且对其进行了扩展。本文介绍的 OpenSL ES 是针对 Android NDK 开发来说。

2. OpenSL ES 的一些基本概念

基于 c 语言的面向对象接口

OpenSL ES 是基于 c 语言实现的，但其提供的接口是采用面向对象的方式实现，OpenSL ES 的大多数 API 是通过对象来调用的。 例如，以下代码片段，主要的逻辑是实例化引擎对象和获取引擎对象接口。

SLresult result;
// realize the engine
result = (*engineObject)->Realize(engineObject, SL_BOOLEAN_FALSE);
assert(SL_RESULT_SUCCESS == result);
(void)result;
result = (*engineObject)->GetInterface(engineObject, SL_IID_ENGINE, &engineEngine);
assert(SL_RESULT_SUCCESS == result);
(void)result;

对象和接口概念

Object 和 Interface OpenSL ES 中的两大基本概念，可以类比为 Java 中的对象和接口 。

在 OpenSL ES 中， 每个 Object 可以存在一系列的 Interface ，并且为每个对象都提供了一系列的基本操作，如 Realize，GetState，Destroy 等。重要的一点，只有通过 GetInterface 方法拿到 Object 的 Interface ，才能使用 Object 提供的功能。

对象的生命周期

OpenSL ES 的 Object 一般有三种状态 ，分别是：UNREALIZED （不可用），REALIZED（可用），SUSPENDED（挂起）。

Object 处于 UNREALIZED （不可用）状态时，系统不会为其分配资源；调用 Realize 方法后便进入 REALIZED（可用）状态，此时对象的各个功能和资源可以正常访问；当系统音频相关的硬件设备被其他进程占用时，OpenSL ES Object 便会进入 SUSPENDED （挂起）状态，随后调用 Resume 方法可使对象重回 REALIZED（可用）状态；当 Object 使用结束后，调用 Destroy 方法释放资源，是对象重回 UNREALIZED （不可用）状态。

3. OpenSL ES 常用的 Object 和 Interface

Audio 引擎对象和接口

Audio 引擎对象和接口，即 Engine Object 和 SLEngineItf Interface 。Engine Object 的主要功能是管理 Audio Engine 的生命周期，提供引擎对象的管理接口。引擎对象的使用方法如下：

SLresult result;
// 创建引擎对象
result = slCreateEngine(&engineObject, 0, NULL, 0, NULL, NULL);
assert(SL_RESULT_SUCCESS == result);
(void)result;
// 实例化
result = (*engineObject)->Realize(engineObject, SL_BOOLEAN_FALSE);
assert(SL_RESULT_SUCCESS == result);
(void)result;
// 获取引擎对象接口
result = (*engineObject)->GetInterface(engineObject, SL_IID_ENGINE, &engineEngine);
assert(SL_RESULT_SUCCESS == result);
(void)result;
// 释放引擎对象的资源
result = (*engineObject)->Destroy(engineObject, SL_BOOLEAN_FALSE);
assert(SL_RESULT_SUCCESS == result);
(void)result;

SLRecordItf 和 SLPlayItf

SLRecordItf 和 SLPlayItf 分别抽象多媒体功能 recorder 和 player ，通过 SLEngineItf 的 CreateAudioPlayer 和 CreateAudioRecorder 方法分别创建 player 和 recorder 对象实例。

// 创建 audio recorder 对象
result = (*engineEngine)->CreateAudioRecorder(engineEngine, &recorderObject , &recSource, &dataSink,
                                                  NUM_RECORDER_EXPLICIT_INTERFACES, iids, required);
// 创建 audio player 对象
SLresult result = (*engineEngine)->CreateAudioPlayer(
        engineEngine,
        &audioPlayerObject,
        &dataSource,
        &dataSink,
        interfaceIDs,
        requiredInterfaces
);

SLDataSource 和 SLDataSink

OpenSL ES 中的 SLDataSource 和 SLDataSink 结构体，主要用于构建 audio player 和 recorder 对象，其中 SLDataSource 表示音频数据来源的信息，SLDataSink 表示音频数据输出信息。

// 数据源简单缓冲队列定位器
SLDataLocator_AndroidSimpleBufferQueue dataSou
        SL_DATALOCATOR_ANDROIDSIMPLEBUFFERQUEU
// 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, //采样率
        wav_get_bits(wav), // 位宽
        wav_get_bits(wav),
        SL_SPEAKER_FRONT_CENTER, // 通道屏蔽
        SL_BYTEORDER_LITTLEENDIAN // 字节顺序
// 数据源
SLDataSource dataSource = {
        &dataSourceLocator,
        &dataSourceFormat
// 针对数据接收器的输出混合定位器(混音器)
SLDataLocator_OutputMix dataSinkLocator = {
        SL_DATALOCATOR_OUTPUTMIX, // 定位器类型
        outputMixObject // 输出混合
// 输出
SLDataSink dataSink = {
        &dataSinkLocator, // 定位器
};

4. OpenSL ES Recorder 和 Player 功能构建

Audio Recorder

Audio Player

PS：Audio Player 的 Data Source 也可以是本地存储或缓存的音频数据，以上图片来自于 Jhuster 的博客。

5. 代码实现

以下代码主要实现音频数据的采集、保存和播放。

#include <jni.h>
#include <string>
#include <assert.h>
#include <SLES/OpenSLES.h>
#include <SLES/OpenSLES_Android.h>
#include <android/log.h>
#define AUDIO_SRC_PATH "/sdcard/audio.pcm"
#define LOGI(FORMAT, ...) __android_log_print(ANDROID_LOG_INFO,"byteflow",FORMAT,##__VA_ARGS__);
#define LOGE(FORMAT, ...) __android_log_print(ANDROID_LOG_ERROR,"byteflow",FORMAT,##__VA_ARGS__);
#define NUM_RECORDER_EXPLICIT_INTERFACES 2
#define NUM_BUFFER_QUEUE 1
#define SAMPLE_RATE 44100
#define PERIOD_TIME 20  // 20ms
#define FRAME_SIZE SAMPLE_RATE * PERIOD_TIME / 1000
#define CHANNELS 2
#define BUFFER_SIZE   (FRAME_SIZE * CHANNELS)
// engine interfaces
static SLObjectItf engineObject = NULL;
static SLEngineItf engineEngine = NULL;
// audio recorder interfaces
static SLObjectItf recorderObject = NULL;
static SLRecordItf recorderRecord = NULL;
static SLAndroidSimpleBufferQueueItf recorderBuffQueueItf = NULL;
static SLAndroidConfigurationItf configItf = NULL;
// pcm audio player interfaces
static SLObjectItf playerObject = NULL;
static SLPlayItf playerPlay = NULL;
static SLObjectItf outputMixObjext = NULL; // 混音器
static SLAndroidSimpleBufferQueueItf playerBufferQueueItf = NULL;
void createEngine(){
    SLEngineOption EngineOption[] = {
            {(SLuint32) SL_ENGINEOPTION_THREADSAFE, (SLuint32) SL_BOOLEAN_TRUE}
    SLresult result;
    result = slCreateEngine(&engineObject, 1, EngineOption, 0, NULL, NULL);
    assert(SL_RESULT_SUCCESS == result);
    /* Realizing the SL Engine in synchronous mode. */
    result = (*engineObject)->Realize(engineObject, SL_BOOLEAN_FALSE);
    assert(SL_RESULT_SUCCESS == result);
    // get the engine interface, which is needed in order to create other objects
    result = (*engineObject)->GetInterface(engineObject, SL_IID_ENGINE, &engineEngine);
    assert(SL_RESULT_SUCCESS == result);
class AudioContext {
public:
    FILE *pfile;
    uint8_t *buffer;
    size_t bufferSize;
    AudioContext(FILE *pfile, uint8_t *buffer, size_t bufferSize){
        this->pfile = pfile;
        this->buffer = buffer;
        this->bufferSize = bufferSize;
static AudioContext *recorderContext = NULL;
// 录制音频时的回调
void AudioRecorderCallback(SLAndroidSimpleBufferQueueItf bufferQueueItf, void *context){
    AudioContext *recorderContext = (AudioContext*)context;
    assert(recorderContext != NULL);
    if (recorderContext->buffer != NULL) {
        fwrite(recorderContext->buffer, recorderContext->bufferSize, 1, recorderContext->pfile);
        LOGI("save a frame audio data.");
        SLresult result;
        SLuint32 state;
        result = (*recorderRecord)->GetRecordState(recorderRecord, &state);
        assert(SL_RESULT_SUCCESS == result);
        (void) result;
        if (state == SL_RECORDSTATE_RECORDING) {
            result = (*bufferQueueItf)->Enqueue(bufferQueueItf, recorderContext->buffer, recorderContext->bufferSize);
            assert(SL_RESULT_SUCCESS == result);
            (void) result;
// 播放音频时的回调
void AudioPlayerCallback(SLAndroidSimpleBufferQueueItf bufferQueueItf, void *context){
    AudioContext *playerContext = (AudioContext*)context;
    if (!feof(playerContext->pfile)) {
        fread(playerContext->buffer, playerContext->bufferSize, 1, playerContext->pfile);
        LOGI("read a frame audio data.");
        (*bufferQueueItf)->Enqueue(bufferQueueItf, playerContext->buffer, playerContext->bufferSize);
    } else {
        fclose(playerContext->pfile);
        delete playerContext->buffer;
// 创建音频播放器
void createAudioPlayer(SLEngineItf engineEngine, SLObjectItf outputMixObject, SLObjectItf &audioPlayerObject){
    SLDataLocator_AndroidSimpleBufferQueue dataSourceLocator = {
            SL_DATALOCATOR_ANDROIDSIMPLEBUFFERQUEUE,
    // PCM 数据源格式
    SLDataFormat_PCM dataSourceFormat = {
            SL_DATAFORMAT_PCM,
            SL_SAMPLINGRATE_44_1,
            SL_PCMSAMPLEFORMAT_FIXED_16,
            SL_SPEAKER_FRONT_LEFT| SL_SPEAKER_FRONT_RIGHT,
            SL_BYTEORDER_LITTLEENDIAN
    SLDataSource dataSource = {
            &dataSourceLocator,
            &dataSourceFormat
    SLDataLocator_OutputMix dataSinkLocator = {
            SL_DATALOCATOR_OUTPUTMIX, // 定位器类型
            outputMixObject // 输出混合
    SLDataSink dataSink = {
            &dataSinkLocator, // 定位器
    // 需要的接口
    SLInterfaceID interfaceIDs[] = {
            SL_IID_BUFFERQUEUE
    SLboolean requiredInterfaces[] = {
            SL_BOOLEAN_TRUE
    // 创建音频播放对象
    SLresult result = (*engineEngine)->CreateAudioPlayer(
            engineEngine,
            &audioPlayerObject,
            &dataSource,
            &dataSink,
            interfaceIDs,
            requiredInterfaces
    assert(SL_RESULT_SUCCESS == result);
    (void) result;
extern "C" {
// 开始播放音频
JNIEXPORT void JNICALL
Java_com_byteflow_opensl_1es_AudioRecorder_startPlay(JNIEnv *env, jobject instance) {
    // 创建引擎
    if (engineEngine == NULL) {
        createEngine();
    // 创建混音器
    SLresult result;
    result = (*engineEngine)->CreateOutputMix(engineEngine, &outputMixObjext, 0, 0, 0);
    assert(SL_RESULT_SUCCESS == result);
    (void) result;
    result = (*outputMixObjext)->Realize(outputMixObjext, SL_BOOLEAN_FALSE);
    assert(SL_RESULT_SUCCESS == result);
    (void) result;
    FILE *p_file = fopen(AUDIO_SRC_PATH, "r");
    // 创建播放器
    createAudioPlayer(engineEngine, outputMixObjext, playerObject);
    result = (*playerObject)->Realize(playerObject, SL_BOOLEAN_FALSE);
    assert(SL_RESULT_SUCCESS == result);
    (void) result;
    result = (*playerObject)->GetInterface(playerObject, SL_IID_BUFFERQUEUE,
                                                &playerBufferQueueItf);
    assert(SL_RESULT_SUCCESS == result);
    (void) result;
    uint8_t *buffer = new uint8_t[BUFFER_SIZE];
    AudioContext *playerContext = new AudioContext(p_file, buffer, BUFFER_SIZE);
    result = (*playerBufferQueueItf)->RegisterCallback(playerBufferQueueItf, AudioPlayerCallback,
                                                    playerContext);
    assert(SL_RESULT_SUCCESS == result);
    (void) result;
    result = (*playerObject)->GetInterface(playerObject, SL_IID_PLAY, &playerPlay);
    assert(SL_RESULT_SUCCESS == result);
    (void) result;
    result = (*playerPlay)->SetPlayState(playerPlay, SL_PLAYSTATE_PLAYING);
    assert(SL_RESULT_SUCCESS == result);
    AudioPlayerCallback(playerBufferQueueItf, playerContext);
// 停止播放音频
JNIEXPORT void JNICALL
Java_com_byteflow_opensl_1es_AudioRecorder_stopPlay(JNIEnv *env, jobject instance) {
    if (playerPlay != NULL) {
        SLresult result;
        result = (*playerPlay)->SetPlayState(playerPlay, SL_PLAYSTATE_STOPPED);
        assert(SL_RESULT_SUCCESS == result);
// 开始采集音频数据，并保存到本地
JNIEXPORT void JNICALL
Java_com_byteflow_opensl_1es_AudioRecorder_startRecord(JNIEnv *env, jobject instance) {
    if (engineEngine == NULL) {
        createEngine();
    if (recorderObject != NULL) {
        LOGI("Audio recorder already has been created.");
        return ;
    FILE *p_file = fopen(AUDIO_SRC_PATH, "w");
    if (p_file == NULL) {
        LOGI("Fail to open file.");
        return ;
    SLresult result;
    /* setup the data source*/
    SLDataLocator_IODevice ioDevice = {
            SL_DATALOCATOR_IODEVICE,
            SL_IODEVICE_AUDIOINPUT,
            SL_DEFAULTDEVICEID_AUDIOINPUT,
    SLDataSource recSource = {&ioDevice, NULL};
    SLDataLocator_AndroidSimpleBufferQueue recBufferQueue = {
            SL_DATALOCATOR_ANDROIDSIMPLEBUFFERQUEUE,
            NUM_BUFFER_QUEUE
    SLDataFormat_PCM pcm = {
            SL_DATAFORMAT_PCM, // pcm 格式的数据
            2,  // 2 个声道（立体声）
            SL_SAMPLINGRATE_44_1, // 44100hz 的采样频率
            SL_PCMSAMPLEFORMAT_FIXED_16,
            SL_PCMSAMPLEFORMAT_FIXED_16,
            SL_SPEAKER_FRONT_LEFT| SL_SPEAKER_FRONT_RIGHT,
            SL_BYTEORDER_LITTLEENDIAN
    SLDataSink dataSink = { &recBufferQueue, &pcm };
    SLInterfaceID iids[NUM_RECORDER_EXPLICIT_INTERFACES] = {SL_IID_ANDROIDSIMPLEBUFFERQUEUE, SL_IID_ANDROIDCONFIGURATION};
    SLboolean required[NUM_RECORDER_EXPLICIT_INTERFACES] = {SL_BOOLEAN_TRUE, SL_BOOLEAN_TRUE};
    /* Create the audio recorder */
    result = (*engineEngine)->CreateAudioRecorder(engineEngine, &recorderObject , &recSource, &dataSink,
                                                  NUM_RECORDER_EXPLICIT_INTERFACES, iids, required);
    assert(SL_RESULT_SUCCESS == result);
    /* get the android configuration interface*/
    result = (*recorderObject)->GetInterface(recorderObject, SL_IID_ANDROIDCONFIGURATION, &configItf);
    assert(SL_RESULT_SUCCESS == result);
    /* Realize the recorder in synchronous mode. */
    result = (*recorderObject)->Realize(recorderObject, SL_BOOLEAN_FALSE);
    assert(SL_RESULT_SUCCESS == result);
    /* Get the buffer queue interface which was explicitly requested */
    result = (*recorderObject)->GetInterface(recorderObject, SL_IID_ANDROIDSIMPLEBUFFERQUEUE, (void*) &recorderBuffQueueItf);
    assert(SL_RESULT_SUCCESS == result);
    /* get the record interface */
    result = (*recorderObject)->GetInterface(recorderObject, SL_IID_RECORD, &recorderRecord);
    assert(SL_RESULT_SUCCESS == result);
    uint8_t *buffer = new uint8_t[BUFFER_SIZE];
    recorderContext = new AudioContext(p_file, buffer, BUFFER_SIZE);
    result = (*recorderBuffQueueItf)->RegisterCallback(recorderBuffQueueItf, AudioRecorderCallback, recorderContext);
    assert(SL_RESULT_SUCCESS == result);
    /* Enqueue buffers to map the region of memory allocated to store the recorded data */
    result = (*recorderBuffQueueItf)->Enqueue(recorderBuffQueueItf, recorderContext->buffer, BUFFER_SIZE);
    assert(SL_RESULT_SUCCESS == result);
    /* Start recording */
    // 开始录制音频
    result = (*recorderRecord)->SetRecordState(recorderRecord, SL_RECORDSTATE_RECORDING);
    assert(SL_RESULT_SUCCESS == result);
    LOGI("Starting recording");
// 停止音频采集
JNIEXPORT void JNICALL
Java_com_byteflow_opensl_1es_AudioRecorder_stopRecord(JNIEnv *env, jobject instance) {
    if (recorderRecord != NULL) {
        SLresult result = (*recorderRecord)->SetRecordState(recorderRecord, SL_RECORDSTATE_STOPPED);
        assert(SL_RESULT_SUCCESS == result);
        if (recorderContext != NULL) {
            fclose(recorderContext->pfile);
            delete recorderContext->buffer;
// 释放资源
JNIEXPORT void JNICALL
Java_com_byteflow_opensl_1es_AudioRecorder_release(JNIEnv *env, jobject instance) {
    if (recorderObject != NULL) {
        (*recorderObject)->Destroy(recorderObject);
        recorderObject = NULL;
        recorderRecord = NULL;
        recorderBuffQueueItf = NULL;
        configItf = NULL;
        recorderContext = NULL;
    if (playerObject != NULL) {
        (*playerObject)->Destroy(playerObject);
        playerObject = NULL;
        playerPlay = NULL;
        playerBufferQueueItf = NULL;
        outputMixObjext = NULL;
    // destroy engine object, and invalidate all associated interfaces
    if (engineObject != NULL) {
        (*engineObject)->Destroy(engineObject);
        engineObject = NULL;
        engineEngine = NULL;
};

CMake 脚本 CMakeLists.txt 。

cmake_minimum_required(VERSION 3.4.1)
add_library( # Sets the name of the library.
             audio-recorder
             # Sets the library as a shared library.
             SHARED
             # Provides a relative path to your source file(s).