音頻重采樣的概念(Audio Resample)

將一個(gè)音頻X，更改X的采樣率、采樣格式、聲道數(shù)等參數(shù)，最終轉(zhuǎn)換成音頻Y，這個(gè)過(guò)程就叫做音頻重采樣
例如X音頻(sampleRate:44100,format:s16le,channel:2)---->Y音頻(sampleRate:48000,format:f32le,channel:1)，其實(shí)只要更改一個(gè)參數(shù)就算是重采樣了。

重采樣.png

上面的流程圖就描述了一個(gè)重采樣的過(guò)程圖，下面讓我們通過(guò)命令行來(lái)操作一下ffmpeg如何重采樣音頻

命令行操控

將44100_s16le_2 轉(zhuǎn)換成4800_f32l3_1

ffmpeg -ar 44100 -ac 2 -f s16le -i 44100_s16le_2.pcm -ar 48000 -ac 1 -f f32le 48000_f32le_1.pcm

在目錄下面可以看到多生成了一個(gè)48000_f32le_1.pcm，這個(gè)就是我們重采樣后的音頻數(shù)據(jù)

播放 48000_f32le_1.pcm

這時(shí)候我們就不能采用原來(lái)的音頻播放格式來(lái)播放這個(gè)采樣后的數(shù)據(jù)，因此播放命令應(yīng)該如下：ffplay -ar 48000 -ac 1 -f 32le   48000_f32le_1.pcm 音頻數(shù)據(jù)正常播放

程序代碼實(shí)現(xiàn)音頻重采樣

下面給出詳細(xì)代碼

• 播放工具類(lèi)ffmpegs.h

#ifndef FFMPEGS_H
#define FFMPEGS_H

extern "C" {
#include <libavformat/avformat.h>
}

typedef struct{
    const char *filename;
    int sampleRate;
    AVSampleFormat sampleFmt;
    int chLayout;
} ResampleAudioSpec;

class FFmpegs
{
public:
    FFmpegs();
    static void resampleAudio(ResampleAudioSpec &in,ResampleAudioSpec &out);
    static void resampleAudio(const char *inFilename,
                              int inSampleRate,
                              AVSampleFormat inSampleFmt,
                              int inChLayout,
                              const char *outFilename,
                              int outSampleRate,
                              AVSampleFormat outSampleFmt,
                              int outChaLayout);
};

#endif // FFMPEGS_H

• 播放工具類(lèi)ffmpegs.cpp

#include "ffmpegs.h"
#include <QDebug>
#include <QFile>

extern "C" {
#include <libswresample/swresample.h>
#include <libavutil/avutil.h>
}

#define ERROR_BUF(ref) \
    char errbuf[1024]; \
    av_strerror(ret,errbuf,sizeof (errbuf));

FFmpegs::FFmpegs()
{

}

void FFmpegs::resampleAudio(ResampleAudioSpec &in, ResampleAudioSpec &out){
    resampleAudio(in.filename,in.sampleRate,in.sampleFmt,in.chLayout,
                  out.filename,out.sampleRate,out.sampleFmt,out.chLayout);
}

void FFmpegs::resampleAudio(const char *inFilename,
                            int inSampleRate,
                            AVSampleFormat inSampleFmt,
                            int inChLayout,
                            const char *outFilename,
                            int outSampleRate,
                            AVSampleFormat outSampleFmt,
                            int outChaLayout){
// 向下取整，AV_ROUND_DOWN(2.66) = 2
// qDebug() << av_rescale_rnd(8, 1, 3, AV_ROUND_DOWN);

// 向上取整，AV_ROUND_UP(1.25) = 2
// qDebug() << av_rescale_rnd(5, 1, 4, AV_ROUND_UP);

    //文件名
    QFile inFile(inFilename);
    QFile outFile(outFilename);

    //輸入緩沖區(qū)
    //指向緩沖區(qū)的指針
    uint8_t **inData = nullptr;
    //緩沖區(qū)的大小
    int inLinesize = 0;
    //聲道數(shù)
    int inChs = av_get_channel_layout_nb_channels(inChLayout);
    //一個(gè)樣本的大小
    int inBytesPerSample = inChs * av_get_bytes_per_sample(inSampleFmt);
    //緩沖區(qū)的樣本數(shù)量
    int inSamples = 1024;
    //讀取文件數(shù)據(jù)的大小
    int len = 0;

    //輸出緩沖區(qū)
    //指向緩沖區(qū)的指針
    uint8_t **outData = nullptr;
    //緩沖區(qū)的大小
    int outLinessize = 0;
    //聲道數(shù)
    int outChs = av_get_channel_layout_nb_channels(outChaLayout);
    //一個(gè)樣本的大小
    int outBytesPerSample = outChs * av_get_bytes_per_sample(outSampleFmt);
    //緩沖區(qū)的樣本數(shù)量
    int outSamples = av_rescale_rnd(outSampleRate,inSamples,inSampleRate,AV_ROUND_UP);
    //計(jì)算公式
    /*
        inSampleRate        inSamples
        ------------- = -------------
        outSampleRate       outSamples
        outSamples = outSampleRate * inSamples / inSampleRate
    */
    qDebug() << "輸入緩沖區(qū)" << inSampleRate << inSamples;
    qDebug() << "輸出緩沖區(qū)" << outSampleRate << outSamples;

      // 返回結(jié)果
    int ret = 0;

    //創(chuàng)建重采樣上下文
    SwrContext *ctx = swr_alloc_set_opts(nullptr,
                                         //取出參數(shù)
                                         outChaLayout,outSampleFmt,outSampleRate,
                                         //輸入?yún)?shù)
                                         inChLayout,inSampleFmt,inSampleRate,
                                         0,nullptr);
    if(!ctx){
        qDebug() << "swr_alloc_set_opts error";
        goto end;
    }

    //初始化重采樣上下文
    ret = swr_init(ctx);
    if(ret < 0){
        ERROR_BUF(ret);
        qDebug() << "swr_init error:" << errbuf;
        goto end;
    }
    /* 指針類(lèi)型（64bit，8個(gè)字節(jié)）
      int *;
      double *;
      void *;
      int **;
      int ***;
      int ******;
      */

      // int *p;
      // *(p + i) == p[i]
      // *(p + 0) == p[0]
      // *p == p[0]

      // int *p = new int[15];
      // int *p = av_calloc(15, sizeof (int));
      // int **pp = av_calloc(7, sizeof (int *));

      // uint8_t **inData = av_calloc(1, sizeof(uint8_t *));

    //創(chuàng)建輸入緩沖區(qū)
    ret = av_samples_alloc_array_and_samples(&inData,
                                             &inLinesize,
                                             inChs,
                                             inSamples,
                                             inSampleFmt,
                                             1);
    if(ret < 0){
        ERROR_BUF(ret);
        qDebug() << "av_samples_alloc_array_and_samples error:" << errbuf;
        goto end;
    }

    //創(chuàng)建輸出緩沖區(qū)
    ret = av_samples_alloc_array_and_samples(
                &outData,
                &outLinessize,
                outChs,
                outSamples,
                outSampleFmt,
                1);
    if (ret < 0) {
            ERROR_BUF(ret);
            qDebug() << "av_samples_alloc_array_and_samples error:" << errbuf;
            goto end;
        }

    //打開(kāi)文件
    if(!inFile.open(QFile::ReadOnly)){
        qDebug() << "file open error:" << outFilename;
        goto end;
    }
    //之前漏掉下面這一句，導(dǎo)致項(xiàng)目一直在報(bào) QIODevice::write (QFile, "/Users/songlin/audio/resample/test/48000_f32le_1.pcm"): device not open
    if (!outFile.open(QFile::WriteOnly)) {
          qDebug() << "file open error:" << outFilename;
          goto end;
      }

    //讀取文件數(shù)據(jù)
    //inData[0] == *inData;
    while((len = inFile.read((char *) inData[0],inLinesize)) > 0){
        //讀取的樣本數(shù)量
        inSamples = len / inBytesPerSample;
        //重采樣(返回值轉(zhuǎn)換后的樣本數(shù)量)
        ret = swr_convert(ctx,outData,outSamples,(const uint8_t **)inData,inSamples);
        if (ret < 0){
            ERROR_BUF(ret);
            qDebug() << "swr_convert error:" << errbuf;
            goto end;
        }

        //        int size = av_samples_get_buffer_size(nullptr, outChs, ret, outSampleFmt, 1);
        //        outFile.write((char *) outData[0], size);

        // 將轉(zhuǎn)換后的數(shù)據(jù)寫(xiě)入到輸出文件中
        // outData[0] == *outData
        outFile.write((char *) outData[0],ret * outBytesPerSample);
    }

end:
    // 釋放資源
    // 關(guān)閉文件
    inFile.close();
    outFile.close();

    // 釋放輸入緩沖區(qū)
    if (inData) {
        av_freep(&inData[0]);
    }
    av_freep(&inData);

    // 釋放輸出緩沖區(qū)
    if (outData) {
        av_freep(&outData[0]);
    }
    av_freep(&outData);

    // 釋放重采樣上下文
    swr_free(&ctx);

//    void *ptr = malloc(100);
//    freep(&ptr);
//    free(ptr);
}

• 子線程audioThread.h

#ifndef AUDIOTHREAD_H
#define AUDIOTHREAD_H

#include <QThread>

class AudioThread : public QThread
{
    Q_OBJECT

private:
    void run();
public:
    explicit AudioThread(QObject *parent = nullptr);
    ~AudioThread();

signals:

};

#endif // AUDIOTHREAD_H

• 子線程audioThread.cpp

#include "audiothread.h"
#include <QDebug>
#include "ffmpegs.h"

AudioThread::AudioThread(QObject *parent) : QThread(parent) {
    // 當(dāng)監(jiān)聽(tīng)到線程結(jié)束時(shí)（finished），就調(diào)用deleteLater回收內(nèi)存
    connect(this, &AudioThread::finished,
            this, &AudioThread::deleteLater);
}


AudioThread::~AudioThread() {
    // 斷開(kāi)所有的連接
    disconnect();
    // 內(nèi)存回收之前，正常結(jié)束線程
    requestInterruption();
    // 安全退出
    quit();
    wait();
    qDebug() << this << "析構(gòu)（內(nèi)存被回收）";
}

void AudioThread::run() {
    // 44100_s16le_2 -> 48000_f32le_2 -> 48000_s32le_1 -> 44100_s16le_2

    ResampleAudioSpec ras1;
    ras1.filename = "/Users/songlin/audio/resample/44100_s16le_2.pcm";
    ras1.sampleFmt = AV_SAMPLE_FMT_S16;
    ras1.sampleRate = 44100;
    ras1.chLayout = AV_CH_LAYOUT_STEREO;

    ResampleAudioSpec ras2;
    ras2.filename = "/Users/songlin/audio/resample/test/48000_f32le_1.pcm";
    ras2.sampleFmt = AV_SAMPLE_FMT_FLT;
    ras2.sampleRate = 48000;
    ras2.chLayout = AV_CH_LAYOUT_MONO;

    ResampleAudioSpec ras3;
    ras3.filename = "/Users/songlin/audio/resample/test/48000_s32le_1.pcm";
    ras3.sampleFmt = AV_SAMPLE_FMT_S32;
    ras3.sampleRate = 48000;
    ras3.chLayout = AV_CH_LAYOUT_MONO;

    ResampleAudioSpec ras4 = ras1;
    ras4.filename = "/Users/songlin/audio/resample/test/44100_s16le_2_new.pcm";

    FFmpegs::resampleAudio(ras1, ras2);
    FFmpegs::resampleAudio(ras2, ras3);
    FFmpegs::resampleAudio(ras3, ras4);
}

• mainwindow.h

#ifndef MAINWINDOW_H
#define MAINWINDOW_H

#include <QMainWindow>
#include <audiothread.h>

QT_BEGIN_NAMESPACE
namespace Ui { class MainWindow; }
QT_END_NAMESPACE

class MainWindow : public QMainWindow
{
    Q_OBJECT

public:
    MainWindow(QWidget *parent = nullptr);
    ~MainWindow();

private slots:
    void on_audioButton_clicked();

private:
    Ui::MainWindow *ui;
    AudioThread *_audioThread = nullptr;
};
#endif // MAINWINDOW_H

• mainwindo.cpp

#include "mainwindow.h"
#include "ui_mainwindow.h"

MainWindow::MainWindow(QWidget *parent)
    : QMainWindow(parent)
    , ui(new Ui::MainWindow)
{
    ui->setupUi(this);
}

MainWindow::~MainWindow()
{
    delete ui;
}


void MainWindow::on_audioButton_clicked()
{
    _audioThread = new AudioThread(this);
    _audioThread->start();
}

到這里，重采樣幾乎已經(jīng)實(shí)現(xiàn)了，代碼中不懂之處，歡迎留言提問(wèn)