：使用ffmpeg进行音频解码

Posted 2022-12-05 薛萌

和视频解码前面的套路一毛一样，麻烦的要对音频参数进行配置

#include "com_xuemeng_mylive_utils_XuemengPlayer.h"
#include <stdlib.h>
#include <stdio.h>
#include <unistd.h>
#include <android/log.h>

#define LOGI(FORMAT,...) __android_log_print(ANDROID_LOG_INFO,"xuemeng",FORMAT,##__VA_ARGS__);
#define LOGE(FORMAT,...) __android_log_print(ANDROID_LOG_ERROR,"xuemeng",FORMAT,##__VA_ARGS__);
#include "libyuv.h"
//封装格式
#include "libavformat/avformat.h"
//解码
#include "libavcodec/avcodec.h"
//缩放
#include "libswscale/swscale.h"

#include "libswresample/swresample.h"

JNIEXPORT void JNICALL Java_com_xuemeng_mylive_utils_XuemengPlayer_sound(JNIEnv *env, jobject jthiz, jstring input_jstr,
        jstring output_jstr) 

    const char* input_cstr = (*env)->GetStringUTFChars(env, input_jstr, NULL);
    const char* output_cstr = (*env)->GetStringUTFChars(env, output_jstr, NULL);
    //1.注册组件
    av_register_all();

    //封装格式上下文
    AVFormatContext *pFormatCtx = avformat_alloc_context();

    //2.打开输入音频文件
    if (avformat_open_input(&pFormatCtx, input_cstr, NULL, NULL) != 0) 
        LOGI("%s", "打开输入音频文件失败");
        return;
    
    //3.获取音频信息
    if (avformat_find_stream_info(pFormatCtx, NULL) < 0) 
        LOGI("%s", "获取音频信息失败");
        return;
    

    //音频解码，需要找到对应的AVStream所在的pFormatCtx->streams的索引位置
    int audio_stream_idx = -1;
    int i = 0;
    for (; i < pFormatCtx->nb_streams; i++) 
        //根据类型判断是否是音频流
        if (pFormatCtx->streams[i]->codec->codec_type == AVMEDIA_TYPE_AUDIO) 
            audio_stream_idx = i;
            break;
        
    
    //4.获取解码器
    //根据索引拿到对应的流,根据流拿到解码器上下文
    AVCodecContext *pCodeCtx = pFormatCtx->streams[audio_stream_idx]->codec;
    //再根据上下文拿到编解码id，通过该id拿到解码器
    AVCodec *pCodec = avcodec_find_decoder(pCodeCtx->codec_id);
    if (pCodec == NULL) 
        LOGI("%s", "无法解码");
        return;
    
    //5.打开解码器
    if (avcodec_open2(pCodeCtx, pCodec, NULL) < 0) 
        LOGI("%s", "编码器无法打开");
        return;
    
    //编码数据
    AVPacket *packet = av_malloc(sizeof(AVPacket));
    //解压缩数据
    AVFrame *frame = av_frame_alloc();

音频这里要进行参数设置

//frame->16bit 44100 PCM 统一音频采样格式与采样率
    SwrContext *swrCtx = swr_alloc();
    //重采样设置选项-----------------------------------------------------------start
    //输入的采样格式
    enum AVSampleFormat in_sample_fmt = pCodeCtx->sample_fmt;
    //输出的采样格式 16bit PCM
    enum AVSampleFormat out_sample_fmt = AV_SAMPLE_FMT_S16;
    //输入的采样率
    int in_sample_rate = pCodeCtx->sample_rate;
    //输出的采样率
    int out_sample_rate = 44100;
    //输入的声道布局
    uint64_t in_ch_layout = pCodeCtx->channel_layout;
    //输出的声道布局
    uint64_t out_ch_layout = AV_CH_LAYOUT_STEREO;

    swr_alloc_set_opts(swrCtx, out_ch_layout, out_sample_fmt, out_sample_rate, in_ch_layout, in_sample_fmt,
            in_sample_rate, 0, NULL);
    swr_init(swrCtx);
    //重采样设置选项-----------------------------------------------------------end
    //获取输出的声道个数
    int out_channel_nb = av_get_channel_layout_nb_channels(out_ch_layout);

---

为了播放音频，使用C调用Java的方式来调用
在我的java层我初始化了一个AudioTrack对象
该方法在 com.xuemeng.mylive.utils.XuemengPlayer类中，为了获得该方法的签名，进入项目的class目录使用javap命令来获取

//java层代码
public AudioTrack createAudioTrack() 

        int sampleRateInHz = 44100;
        int audioFormat = AudioFormat.ENCODING_PCM_16BIT;
        int channelConfig = android.media.AudioFormat.CHANNEL_OUT_STEREO;
        int bufferSizeInBytes = AudioTrack.getMinBufferSize(sampleRateInHz,
                channelConfig, audioFormat);
        AudioTrack audioTrack = new AudioTrack(AudioManager.STREAM_MUSIC,
                sampleRateInHz, channelConfig, audioFormat, bufferSizeInBytes,
                AudioTrack.MODE_STREAM);

        // audioTrack.write(audioData, offsetInBytes, sizeInBytes);
        return audioTrack;
    

---

//JNI调用-----------------------------------------------------------------start
    //XuemengPlayer
    jclass player_class = (*env)->GetObjectClass(env, jthiz);
    //AudioTrack对象
    jmethodID create_audio_track_mid = (*env)->GetMethodID(env, player_class, "createAudioTrack",
            "()Landroid/media/AudioTrack;");
    jobject audio_track = (*env)->CallObjectMethod(env, jthiz, create_audio_track_mid);
    //调用AudioTrack.play方法
    jclass audio_track_class = (*env)->GetObjectClass(env, audio_track);
    jmethodID audio_track_play_mid = (*env)->GetMethodID(env, audio_track_class, "play", "()V");
    (*env)->CallVoidMethod(env, audio_track, audio_track_play_mid);
    //AudioTrack.write
    jmethodID audio_track_write_mid = (*env)->GetMethodID(env, audio_track_class, "write", "([BII)I");

    //JNI调用-----------------------------------------------------------------end

解码音频使用avcodec_decode_audio4方法

//存储pcm数据
    uint8_t *out_buffer = (uint8_t *) av_malloc(2 * 44100);
    //FILE *fp_pcm = fopen(output_cstr, "wb");
    int ret, got_frame, framecount = 0;
    //6.一帧一帧读取压缩的音频数据AVPacket
    while (av_read_frame(pFormatCtx, packet) >= 0) 
        if (packet->stream_index == audio_stream_idx) 
            //解码AVPacket->AVFrame
            ret = avcodec_decode_audio4(pCodeCtx, frame, &got_frame, packet);
            if (ret < 0) 
                LOGI("%s", "解码完成");
            
            //非0，正在解码
            if (got_frame) 
                LOGI("解码%d帧", framecount++);
                swr_convert(swrCtx, &out_buffer, 2 * 44100, frame->data, frame->nb_samples);
                //获取sample的size
                int out_buffer_size = av_samples_get_buffer_size(NULL, out_channel_nb, frame->nb_samples,
                        out_sample_fmt, 1);
                //写入文件进行测试
                //fwrite(out_buffer, 1, out_buffer_size, fp_pcm);
                //out_buffer缓冲区数据转byte数组
                jbyteArray audio_sample_array = (*env)->NewByteArray(env, out_buffer_size);
                jbyte *sample_bytep = (*env)->GetByteArrayElements(env, audio_sample_array, NULL);
                //out_buffer数据复制到sample_bytep
                memcpy(sample_bytep, out_buffer, out_buffer_size);
                //同步
                (*env)->ReleaseByteArrayElements(env, audio_sample_array, sample_bytep, 0);
                //AudioTrack.write PCM数据
                (*env)->CallIntMethod(env, audio_track, audio_track_write_mid, audio_sample_array, 0, out_buffer_size);
                //释放局部引用
                (*env)->DeleteLocalRef(env, audio_sample_array);
                usleep(1000 * 16);
            
        
        av_free_packet(packet);
    

//fclose(fp_pcm);
    av_frame_free(&frame);
    av_free(out_buffer);
    swr_free(&swrCtx);
    avcodec_close(pCodeCtx);
    avformat_close_input(&pFormatCtx);

    (*env)->ReleaseStringUTFChars(env, input_jstr, input_cstr);
    (*env)->ReleaseStringUTFChars(env, output_jstr, output_cstr);