使用FFMPEG进行音频重采样

准备

1. ffmpeg 4.4

2. sdl2

3.一段原始的音频PCM数据

重采样流程

1.设置输入音频参数和输出音频参数

2.根据设置的参数初始化SwrContent上下文

3.创建一个输入buffer, 根据输入的音频参数（采样率，通道数，样本位深度）申请空间，填入默认数据，用于存储输入音频数据

4.创建一个输出buffer, 根据输出的音频参数（采样率，通道数，样本位深度）申请空间，填入默认数据，用于存储重采样后的数据

5.读取PCM数据，每次读取的大小等于输入buffer的大小

6.进行重采样swr_convert

7.将输出的buffer拷贝到SDL2音频回调缓冲区中播放，或者直接写入文件，使用ffplay进行测试，也可以封装成Frame送到音频编码器中（如aac），进行编码后保存。

关键代码

设置重采样参数并初始化SWr_Content结构

struct SwrContext* swr_ctx;

swr_ctx = swr_alloc_set_opts(nullptr,
		AV_CH_LAYOUT_MONO, //输出通道
		AV_SAMPLE_FMT_S16, //输出样本格式
		44100, //输出采样率
		AV_CH_LAYOUT_STEREO,  //输入通道
		AV_SAMPLE_FMT_FLT,  //输入样本格式
		44100, //输入采样率
		0, nullptr); 

	swr_init(swr_ctx);

输入/输出buffer 创建

	//输入数据buffer
	uint8_t** pcm_buffer;
	int src_linesize;
	int src_nb_channels = av_get_channel_layout_nb_channels(AV_CH_LAYOUT_STEREO);
	int ret = av_samples_alloc_array_and_samples(&pcm_buffer, &src_linesize, src_nb_channels, frame_nb_samples, AV_SAMPLE_FMT_FLT, 0);
	if (ret < 0) {
		fprintf(stderr, "Could not allocate source samples\n");	
	}

	//输出数据buffer
	uint8_t** out_buffer;
	int dst_linesize;
	int dst_nb_channels = av_get_channel_layout_nb_channels(AV_CH_LAYOUT_MONO);
	ret = av_samples_alloc_array_and_samples(&out_buffer, &dst_linesize, dst_nb_channels, frame_nb_samples, AV_SAMPLE_FMT_S16, 0);
	if (ret < 0) {
		fprintf(stderr, "Could not allocate source samples\n");
	}

读文件并进行重采样

readcount = fread((char *)pcm_buffer[0], 1, src_linesize, fp);

data_count += readcount;
printf("   Now Playing %10d KBytes data.  %d \n", data_count / 1024, readcount);
		
swr_convert(swr_ctx, out_buffer, frame_nb_samples, (const uint8_t**)pcm_buffer, frame_nb_samples);

源码分享

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

#include "sdl.h"

extern "C"
{
	#include <libavutil/opt.h>
	#include <libavutil/channel_layout.h>
	#include <libavutil/samplefmt.h>
	#include <libswresample/swresample.h>
}

static  Uint32  audio_len;
static  Uint8* audio_pos;
int frame_nb_samples = 1024; //一帧数据样本数
struct SwrContext* swr_ctx;


void  fill_audio_pcm(void* udata, Uint8* stream, int len) 
{
	SDL_memset(stream, 0, len);

	if (audio_len == 0)
		return;
	len = (len > audio_len ? audio_len : len);

	SDL_MixAudio(stream, audio_pos, len, SDL_MIX_MAXVOLUME);
	audio_pos += len;
	audio_len -= len;
}

int main(int argc, char* argv[])
{
	if (SDL_Init(SDL_INIT_AUDIO || SDL_INIT_TIMER))
	{
		printf("SDL init error\n");
		return -1;
	}

	swr_ctx = swr_alloc_set_opts(nullptr,
		AV_CH_LAYOUT_MONO, //输出通道
		AV_SAMPLE_FMT_S16, //输出样本格式
		44100, //输出采样率
		AV_CH_LAYOUT_STEREO,  //输入通道
		AV_SAMPLE_FMT_FLT,  //输入样本格式
		44100, //输入采样率
		0, nullptr); 

	swr_init(swr_ctx);

	//SDL_AudioSpec
	SDL_AudioSpec wanted_spec;
	wanted_spec.freq = 44100;
	wanted_spec.format = AUDIO_F32; //AUDIO_S16LSB; //AUDIO_F32;
	wanted_spec.channels = 2;
	wanted_spec.silence = 0;
	wanted_spec.samples = frame_nb_samples;
	wanted_spec.callback = fill_audio_pcm;

	if (SDL_OpenAudio(&wanted_spec, NULL) < 0) {
		printf("can't open audio.\n");
		return -1;
	}
	//Play
	SDL_PauseAudio(0);
	
	//输入数据buffer
	uint8_t** pcm_buffer;
	int src_linesize;
	int src_nb_channels = av_get_channel_layout_nb_channels(AV_CH_LAYOUT_STEREO);
	int ret = av_samples_alloc_array_and_samples(&pcm_buffer, &src_linesize, src_nb_channels, frame_nb_samples, AV_SAMPLE_FMT_FLT, 0);
	if (ret < 0) {
		fprintf(stderr, "Could not allocate source samples\n");	
	}

	//输出数据buffer
	uint8_t** out_buffer;
	int dst_linesize;
	int dst_nb_channels = av_get_channel_layout_nb_channels(AV_CH_LAYOUT_MONO);
	ret = av_samples_alloc_array_and_samples(&out_buffer, &dst_linesize, dst_nb_channels, frame_nb_samples, AV_SAMPLE_FMT_S16, 0);
	if (ret < 0) {
		fprintf(stderr, "Could not allocate source samples\n");
	}

	FILE* fp = nullptr;
	fopen_s(&fp, "D:/工程/音视频分析/source/f32le.pcm", "rb+");
	if (fp == NULL) {
		printf("cannot open this file\n");
		return -1;
	}
	int readcount = -1;
	int data_count = 0;
	while (!feof(fp)) 
	{
		readcount = fread((char *)pcm_buffer[0], 1, src_linesize, fp);

		data_count += readcount;
		printf("   Now Playing %10d KBytes data.  %d \n", data_count / 1024, readcount);
		
		swr_convert(swr_ctx, out_buffer, frame_nb_samples, (const uint8_t**)pcm_buffer, frame_nb_samples);

		//Set audio buffer (PCM data)
		audio_len = dst_linesize; 
		audio_pos =  (Uint8*)out_buffer[0]; //(Uint8*)pcm_buffer[0];

		while (audio_len > 0)
			SDL_Delay(1);
		
	}

	if (pcm_buffer)
		av_freep(&pcm_buffer[0]);
	av_freep(&pcm_buffer);

	if (out_buffer)
		av_freep(&out_buffer[0]);
	av_freep(&out_buffer);

	swr_free(&swr_ctx);

	fclose(fp);
	SDL_Quit();

	return 0;
}