利用ffmpeg的filter混音

最新推荐文章于 2024-03-11 23:45:38 发布
最新推荐文章于 2024-03-11 23:45:38 发布
阅读量1w 收藏 14
点赞数
分类专栏： ffmpeg 文章标签： ffmpeg filter
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/dancing_night/article/details/53080385
版权

概述

一个简单的混音demo，把文件a和文件b的音频混为一个音频流输出并存为文件，只处理每个文件的第一个音频流
注：源代码是网友Larry_Liang(1085803139)写的，我帮其调试通过，存在这里以免备忘

流程介绍

流程如下：

打开输入1->打开输入2->打开输出->初始化混音filter->开启采集线程和编码写文件循环

线程之间数据走向如下：

采集线程1采集到数据->解码->写入fifo1 |
采集线程2采集到数据->解码->写入fifo2 | ->主线程编码循环读取2个fifo->把数据压人filter->从filter中读取混音后数据->编码->写入文件

代码

/*
*一个简单的混音demo，把文件a和文件b的音频混为一个音频流输出并存为文件，只处理每个文件的第一个音频流
*源代码是网友Larry_Liang(1085803139)写的，我帮其调试通过
*MK(821486004@qq.com)
*/


extern "C"
{
#include "libavcodec/avcodec.h"
#include "libavformat/avformat.h"
#include "libavdevice/avdevice.h"
#include "libavfilter/avfilter.h"
#include "libavfilter/avfiltergraph.h"
#include "libavfilter/buffersink.h"
#include "libavfilter/buffersrc.h"
#include "libavutil/audio_fifo.h"
#include "libavutil/avutil.h"
#include "libavutil/fifo.h"
}

#pragma comment(lib, "avcodec.lib")
#pragma comment(lib, "avformat.lib")
#pragma comment(lib, "avutil.lib")
#pragma comment(lib, "avdevice.lib")
#pragma comment(lib, "avfilter.lib")

//#pragma comment(lib, "avfilter.lib")
//#pragma comment(lib, "postproc.lib")
//#pragma comment(lib, "swresample.lib")
#pragma comment(lib, "swscale.lib")

#include <windows.h>
#include <conio.h>
#include <time.h>

enum CaptureState
{
    PREPARED,
    RUNNING,
    STOPPED,
    FINISHED
};

 typedef struct BufferSourceContext {
    const AVClass    *bscclass;
    AVFifoBuffer     *fifo;
    AVRational        time_base;     ///< time_base to set in the output link
    AVRational        frame_rate;    ///< frame_rate to set in the output link
    unsigned          nb_failed_requests;
    unsigned          warning_limit;

    /* video only */
    int               w, h;
    enum AVPixelFormat  pix_fmt;
    AVRational        pixel_aspect;
    char              *sws_param;

    AVBufferRef *hw_frames_ctx;

    /* audio only */
    int sample_rate;
    enum AVSampleFormat sample_fmt;
    int channels;
    uint64_t channel_layout;
    char    *channel_layout_str;

    int got_format_from_params;
    int eof;
 } BufferSourceContext;

AVFormatContext* _fmt_ctx_spk = NULL;
AVFormatContext* _fmt_ctx_mic = NULL;
AVFormatContext* _fmt_ctx_out = NULL;
int _index_spk = -1;
int _index_mic = -1;
int _index_a_out = -1;

AVFilterGraph* _filter_graph = NULL;
AVFilterContext* _filter_ctx_src_spk = NULL;
AVFilterContext* _filter_ctx_src_mic = NULL;
AVFilterContext* _filter_ctx_sink = NULL;

CaptureState _state = CaptureState::PREPARED;

CRITICAL_SECTION _section_spk;
CRITICAL_SECTION _section_mic;
AVAudioFifo* _fifo_spk = NULL;
AVAudioFifo* _fifo_mic = NULL;

void InitRecorder()
{
    av_register_all();
    avdevice_register_all();
    avfilter_register_all();
}

int OpenSpeakerInput(char* inputForamt, char* url)
{
    AVInputFormat* ifmt = av_find_input_format(inputForamt);
    AVDictionary* opt1 = NULL;
    av_dict_set(&opt1, "rtbufsize", "10M", 0);

    int ret = 0;
    ret = avformat_open_input(&_fmt_ctx_spk, url, ifmt, &opt1);
    if (ret < 0)
    {
        printf("Speaker: failed to call avformat_open_input\n");
        return -1;
    }
    ret = avformat_find_stream_info(_fmt_ctx_spk, NULL);
    if (ret < 0)
    {
        printf("Speaker: failed to call avformat_find_stream_info\n");
        return -1;
    }
    for (int i = 0; i < _fmt_ctx_spk->nb_streams; i++)
    {
        if (_fmt_ctx_spk->streams[i]->codec->codec_type == AVMEDIA_TYPE_AUDIO)
        {
            _index_spk = i;
            break;
        }
    }
    if (_index_spk < 0)
    {
        printf("Speaker: negative audio index\n");
        return -1;
    }
    AVCodecContext* codec_ctx = _fmt_ctx_spk->streams[_index_spk]->codec;
    AVCodec* codec = avcodec_find_decoder(codec_ctx->codec_id);
    if (codec == NULL)
    {
        printf("Speaker: null audio decoder\n");
        return -1;
    }
    ret = avcodec_open2(codec_ctx, codec, NULL);
    if (ret < 0)
    {
        printf("Speaker: failed to call avcodec_open2\n");
        return -1;
    }
    av_dump_format(_fmt_ctx_spk, _index_spk, url, 0);

    return 0;
}

int OpenMicrophoneInput(char* inputForamt, char* url)
{
    AVInputFormat* ifmt = av_find_input_format(inputForamt);
    AVDictionary* opt1 = NULL;
    av_dict_set(&opt1, "rtbufsize", "10M", 0);

    int ret = 0;
    ret = avformat_open_input(&_fmt_ctx_mic, url, ifmt, &opt1);
    if (ret < 0)
    {
        printf("Microphone: failed to call avformat_open_input\n");
        return -1;
    }
    ret = avformat_find_stream_info(_fmt_ctx_mic, NULL);
    if (ret < 0)
    {
        printf("Microphone: failed to call avformat_find_stream_info\n");
        return -1;
    }
    for (int i = 0; i < _fmt_ctx_mic->nb_streams; i++)
    {
        if (_fmt_ctx_mic->streams[i]->codec->codec_type == AVMEDIA_TYPE_AUDIO)
        {
            _index_mic = i;
            break;
        }
    }
    if (_index_mic < 0)
    {
        printf("Microphone: negative audio index\n");
        return -1;
    }
    AVCodecContext* codec_ctx = _fmt_ctx_mic->streams[_index_mic]->codec;
    AVCodec* codec = avcodec_find_decoder(codec_ctx->codec_id);
    if (codec == NULL)
    {
        printf("Microphone: null audio decoder\n");
        return -1;
    }
    ret = avcodec_open2(codec_ctx, codec, NULL);
    if (ret < 0)
    {
        printf("Microphone: failed to call avcodec_open2\n");
        return -1;
    }

    av_dump_format(_fmt_ctx_mic, _index_mic, url, 0);

    return 0;
}

int OpenFileOutput(char* fileName)
{
    int ret = 0;
    ret = avformat_alloc_output_context2(&_fmt_ctx_out, NULL, NULL, fileName);
    if (ret < 0)
    {
        printf("Mixer: failed to call avformat_alloc_output_context2\n");
        return -1;
    }
    AVStream* stream_a = NULL;
    stream_a = avformat_new_stream(_fmt_ctx_out, NULL);
    if (stream_a == NULL)
    {
        printf("Mixer: failed to call avformat_new_stream\n");
        return -1;
    }
    _index_a_out = 0;

    stream_a->codec->codec_type = AVMEDIA_TYPE_AUDIO;
    AVCodec* codec_mp3 = avcodec_find_encoder(AV_CODEC_ID_MP3);
    stream_a->codec->codec = codec_mp3;
    stream_a->codec->sample_rate = 44100;
    stream_a->codec->channels = 2;
    stream_a->codec->channel_layout = av_get_default_channel_layout(2);
    stream_a->codec->sample_fmt = codec_mp3->sample_fmts[0];
    stream_a->codec->bit_rate = 320000;
    stream_a->codec->time_base.num = 1;
    stream_a->codec->time_base.den = stream_a->codec->sample_rate;
    stream_a->codec->codec_tag = 0;


    if (_fmt_ctx_out->oformat->flags & AVFMT_GLOBALHEADER)
        stream_a->codec->flags |= CODEC_FLAG_GLOBAL_HEADER;

    if (avcodec_open2(stream_a->codec, stream_a->codec->codec, NULL) < 0)
    {
        printf("Mixer: failed to call avcodec_open2\n");
        return -1; 
    }
    if (!(_fmt_ctx_out->oformat->flags & AVFMT_NOFILE))
    {
        if (avio_open(&_fmt_ctx_out->pb, fileName, AVIO_FLAG_WRITE) < 0)
        {
            printf("Mixer: failed to call avio_open\n");
            return -1;
        }
    }

    if (avformat_write_header(_fmt_ctx_out, NULL) < 0)
    {
        printf("Mixer: failed to call avformat_write_header\n");
        return -1;
    }

    bool b = (!_fmt_ctx_out->streams[0]->time_base.num && _fmt_ctx_out->streams[0]->codec->time_base.num);

    av_dump_format(_fmt_ctx_out, _index_a_out, fileName, 1);

    _fifo_spk = av_audio_fifo_alloc(_fmt_ctx_spk->streams[_index_spk]->codec->sample_fmt, _fmt_ctx_spk->streams[_index_spk]->codec->channels, 30*_fmt_ctx_spk->streams[_index_spk]->codec->frame_size);
    _fifo_mic = av_audio_fifo_alloc(_fmt_ctx_mic->streams[_index_mic]->codec->sample_fmt, _fmt_ctx_mic->streams[_index_mic]->codec->channels, 30*_fmt_ctx_mic->streams[_index_spk]->codec->frame_size);

    return 0;
}

int InitFilter(char* filter_desc)
{
    char args_spk[512];
    char* pad_name_spk = "in0";
    char args_mic[512];
    char* pad_name_mic = "in1";

    AVFilter* filter_src_spk = avfilter_get_by_name("abuffer");
    AVFilter* filter_src_mic = avfilter_get_by_name("abuffer");
    AVFilter* filter_sink = avfilter_get_by_name("abuffersink");
    AVFilterInOut* filter_output_spk = avfilter_inout_alloc();
    AVFilterInOut* filter_output_mic = avfilter_inout_alloc();
    AVFilterInOut* filter_input = avfilter_inout_alloc();
    _filter_graph = avfilter_graph_alloc();

    sprintf_s(args_spk, sizeof(args_spk), "time_base=%d/%d:sample_rate=%d:sample_fmt=%s:channel_layout=0x%I64x", 
        _fmt_ctx_spk->streams[_index_spk]->codec->time_base.num, 
        _fmt_ctx_spk->streams[_index_spk]->codec->time_base.den, 
        _fmt_ctx_spk->streams[_index_spk]->codec->sample_rate, 
        av_get_sample_fmt_name(_fmt_ctx_spk->streams[_index_spk]->codec->sample_fmt), 
        _fmt_ctx_spk->streams[_index_spk]->codec->channel_layout);
    sprintf_s(args_mic, sizeof(args_mic), "time_base=%d/%d:sample_rate=%d:sample_fmt=%s:channel_layout=0x%I64x", 
        _fmt_ctx_mic->streams[_index_mic]->codec->time_base.num, 
        _fmt_ctx_mic->streams[_index_mic]->codec->time_base.den, 
        _fmt_ctx_mic->streams[_index_mic]->codec->sample_rate, 
        av_get_sample_fmt_name(_fmt_ctx_mic->streams[_index_mic]->codec->sample_fmt), 
        _fmt_ctx_mic->streams[_index_mic]->codec->channel_layout);

    //sprintf_s(args_spk, sizeof(args_spk), "time_base=%d/%d:sample_rate=%d:sample_fmt=%s:channel_layout=0x%I64x", _fmt_ctx_out->streams[_index_a_out]->codec->time_base.num, _fmt_ctx_out->streams[_index_a_out]->codec->time_base.den, _fmt_ctx_out->streams[_index_a_out]->codec->sample_rate, av_get_sample_fmt_name(_fmt_ctx_out->streams[_index_a_out]->codec->sample_fmt), _fmt_ctx_out->streams[_index_a_out]->codec->channel_layout);
    //sprintf_s(args_mic, sizeof(args_mic), "time_base=%d/%d:sample_rate=%d:sample_fmt=%s:channel_layout=0x%I64x", _fmt_ctx_out->streams[_index_a_out]->codec->time_base.num, _fmt_ctx_out->streams[_index_a_out]->codec->time_base.den, _fmt_ctx_out->streams[_index_a_out]->codec->sample_rate, av_get_sample_fmt_name(_fmt_ctx_out->streams[_index_a_out]->codec->sample_fmt), _fmt_ctx_out->streams[_index_a_out]->codec->channel_layout);


    int ret = 0;
    ret = avfilter_graph_create_filter(&_filter_ctx_src_spk, filter_src_spk, pad_name_spk, args_spk, NULL, _filter_graph);
    if (ret < 0)
    {
        printf("Filter: failed to call avfilter_graph_create_filter -- src spk\n");
        return -1;
    }
    ret = avfilter_graph_create_filter(&_filter_ctx_src_mic, filter_src_mic, pad_name_mic, args_mic, NULL, _filter_graph);
    if (ret < 0)
    {
        printf("Filter: failed to call avfilter_graph_create_filter -- src mic\n");
        return -1;
    }

    ret = avfilter_graph_create_filter(&_filter_ctx_sink, filter_sink, "out", NULL, NULL, _filter_graph);
    if (ret < 0)
    {
        printf("Filter: failed to call avfilter_graph_create_filter -- sink\n");
        return -1;
    }
    AVCodecContext* encodec_ctx = _fmt_ctx_out->streams[_index_a_out]->codec;
    ret = av_opt_set_bin(_filter_ctx_sink, "sample_fmts", (uint8_t*)&encodec_ctx->sample_fmt, sizeof(encodec_ctx->sample_fmt), AV_OPT_SEARCH_CHILDREN);
    if (ret < 0)
    {
        printf("Filter: failed to call av_opt_set_bin -- sample_fmts\n");
        return -1;
    }
    ret = av_opt_set_bin(_filter_ctx_sink, "channel_layouts", (uint8_t*)&encodec_ctx->channel_layout, sizeof(encodec_ctx->channel_layout), AV_OPT_SEARCH_CHILDREN);
    if (ret < 0)
    {
        printf("Filter: failed to call av_opt_set_bin -- channel_layouts\n");
        return -1;
    }
    ret = av_opt_set_bin(_filter_ctx_sink, "sample_rates", (uint8_t*)&encodec_ctx->sample_rate, sizeof(encodec_ctx->sample_rate), AV_OPT_SEARCH_CHILDREN);
    if (ret < 0)
    {
        printf("Filter: failed to call av_opt_set_bin -- sample_rates\n");
        return -1;
    }

    filter_output_spk->name = av_strdup(pad_name_spk);
    filter_output_spk->filter_ctx = _filter_ctx_src_spk;
    filter_output_spk->pad_idx = 0;
    filter_output_spk->next = filter_output_mic;

    filter_output_mic->name = av_strdup(pad_name_mic);
    filter_output_mic->filter_ctx = _filter_ctx_src_mic;
    filter_output_mic->pad_idx = 0;
    filter_output_mic->next = NULL;

    filter_input->name = av_strdup("out");
    filter_input->filter_ctx = _filter_ctx_sink;
    filter_input->pad_idx = 0;
    filter_input->next = NULL;

    AVFilterInOut* filter_outputs[2];
    filter_outputs[0] = filter_output_spk;
    filter_outputs[1] = filter_output_mic;

    ret = avfilter_graph_parse_ptr(_filter_graph, filter_desc, &filter_input, filter_outputs, NULL);
    if (ret < 0)
    {
        printf("Filter: failed to call avfilter_graph_parse_ptr\n");
        return -1;
    }

    ret = avfilter_graph_config(_filter_graph, NULL);
    if (ret < 0)
    {
        printf("Filter: failed to call avfilter_graph_config\n");
        return -1;
    }

    avfilter_inout_free(&filter_input);
    av_free(filter_src_spk);
    av_free(filter_src_mic);
    avfilter_inout_free(filter_outputs);
    //av_free(filter_outputs);

    char* temp = avfilter_graph_dump(_filter_graph, NULL);
    printf("%s\n", temp);

    return 0;
}

DWORD WINAPI SpeakerCapThreadProc(LPVOID lpParam)
{
    AVFrame* pFrame = av_frame_alloc();
    AVPacket packet;
    av_init_packet(&packet);

    int got_sound;

    while (_state == CaptureState::RUNNING)
    {
        packet.data = NULL;
        packet.size = 0;

        if (av_read_frame(_fmt_ctx_spk, &packet) < 0)
        {
            continue;
        }
        if (packet.stream_index == _index_spk)
        {
            if (avcodec_decode_audio4(_fmt_ctx_spk->streams[_index_spk]->codec, pFrame, &got_sound, &packet) < 0)
            {
                break;
            }
            av_free_packet(&packet);

            if (!got_sound)
            {
                continue;
            }

            int fifo_spk_space = av_audio_fifo_space(_fifo_spk);
            while(fifo_spk_space < pFrame->nb_samples && _state == CaptureState::RUNNING)
            {
                Sleep(10);
                printf("_fifo_spk full !\n");
                fifo_spk_space = av_audio_fifo_space(_fifo_spk);
            }

            if (fifo_spk_space >= pFrame->nb_samples)
            {
                EnterCriticalSection(&_section_spk);
                int nWritten = av_audio_fifo_write(_fifo_spk, (void**)pFrame->data, pFrame->nb_samples);
                LeaveCriticalSection(&_section_spk);
            }
        }
    }
    av_frame_free(&pFrame);

    return 0;
}

DWORD WINAPI MicrophoneCapThreadProc(LPVOID lpParam)
{
    AVFrame* pFrame = av_frame_alloc();
    AVPacket packet;
    av_init_packet(&packet);

    int got_sound;

    while (_state == CaptureState::PREPARED)
    {

    }

    while (_state == CaptureState::RUNNING)
    {
        packet.data = NULL;
        packet.size = 0;

        if (av_read_frame(_fmt_ctx_mic, &packet) < 0)
        {
            continue;
        }
        if (packet.stream_index == _index_mic)
        {
            if (avcodec_decode_audio4(_fmt_ctx_mic->streams[_index_mic]->codec, pFrame, &got_sound, &packet) < 0)
            {
                break;
            }
            av_free_packet(&packet);

            if (!got_sound)
            {
                continue;
            }

            int fifo_mic_space = av_audio_fifo_space(_fifo_mic);
            while(fifo_mic_space < pFrame->nb_samples && _state == CaptureState::RUNNING)
            {
                Sleep(10);
                printf("_fifo_mic full !\n");
                fifo_mic_space = av_audio_fifo_space(_fifo_mic);
            }

            if (fifo_mic_space >= pFrame->nb_samples)
            {
                EnterCriticalSection(&_section_mic);
                int temp = av_audio_fifo_space(_fifo_mic);
                int temp2 = pFrame->nb_samples;
                int nWritten = av_audio_fifo_write(_fifo_mic, (void**)pFrame->data, pFrame->nb_samples);
                LeaveCriticalSection(&_section_mic);
            }
        }
    }
    av_frame_free(&pFrame);

    return 0;
}

int main()
{
    int ret = 0;

    InitRecorder();

    char fileName[128];
    char* outFileType = ".mp3";

    time_t rawtime;
    tm* timeInfo;
    time(&rawtime);
    timeInfo = localtime(&rawtime);
    sprintf_s(fileName, sizeof(fileName), "%d_%d_%d_%d_%d_%d%s",
        timeInfo->tm_year + 1900, timeInfo->tm_mon + 1, timeInfo->tm_mday,
        timeInfo->tm_hour, timeInfo->tm_min, timeInfo->tm_sec, outFileType);

    char* filter_desc = "[in0][in1]amix=inputs=2[out]";

    //ret = OpenSpeakerInput("dshow", "audio=virtual-audio-capturer");
    ret = OpenSpeakerInput(NULL, "故乡.mp3");
    if (ret < 0)
    {
        goto Release;
    }
    //ret = OpenMicrophoneInput("dshow", "audio=External Microphone (Conexant S");
    ret = OpenMicrophoneInput(NULL, "旅行.mp3");
    if (ret < 0)
    {
        goto Release;
    }
    ret = OpenFileOutput(fileName);
    if (ret < 0)
    {
        goto Release;
    }
    ret = InitFilter(filter_desc);
    if (ret < 0)
    {
        goto Release;
    }

    _state = CaptureState::RUNNING;

    InitializeCriticalSection(&_section_spk);
    InitializeCriticalSection(&_section_mic);

    CreateThread(NULL, 0, SpeakerCapThreadProc, 0, 0, NULL);
    CreateThread(NULL, 0, MicrophoneCapThreadProc, 0, 0, NULL);

    int tmpFifoFailed = 0;
    int64_t frame_count = 0;

    while (_state != CaptureState::FINISHED)
    {
        if (_kbhit())
        {
            _state = CaptureState::STOPPED;
            break;
        }
        else
        {
            int ret = 0;
            AVFrame* pFrame_spk = av_frame_alloc();
            AVFrame* pFrame_mic = av_frame_alloc();


            AVPacket packet_out;

            int got_packet_ptr = 0;

            int fifo_spk_size = av_audio_fifo_size(_fifo_spk);
            int fifo_mic_size = av_audio_fifo_size(_fifo_mic);
            int frame_spk_min_size = _fmt_ctx_spk->streams[_index_spk]->codec->frame_size;
            int frame_mic_min_size = _fmt_ctx_mic->streams[_index_mic]->codec->frame_size;
            if (fifo_spk_size >= frame_spk_min_size && fifo_mic_size >= frame_mic_min_size)
            {
                tmpFifoFailed = 0;

                pFrame_spk->nb_samples = frame_spk_min_size;
                pFrame_spk->channel_layout = _fmt_ctx_spk->streams[_index_spk]->codec->channel_layout;
                pFrame_spk->format = _fmt_ctx_spk->streams[_index_spk]->codec->sample_fmt;
                pFrame_spk->sample_rate = _fmt_ctx_spk->streams[_index_spk]->codec->sample_rate;
                av_frame_get_buffer(pFrame_spk, 0);

                pFrame_mic->nb_samples = frame_mic_min_size;
                pFrame_mic->channel_layout = _fmt_ctx_mic->streams[_index_mic]->codec->channel_layout;
                pFrame_mic->format = _fmt_ctx_mic->streams[_index_mic]->codec->sample_fmt;
                pFrame_mic->sample_rate = _fmt_ctx_mic->streams[_index_mic]->codec->sample_rate;
                av_frame_get_buffer(pFrame_mic, 0);

                EnterCriticalSection(&_section_spk);
                ret = av_audio_fifo_read(_fifo_spk, (void**)pFrame_spk->data, frame_spk_min_size);
                LeaveCriticalSection(&_section_spk);

                EnterCriticalSection(&_section_mic);
                ret = av_audio_fifo_read(_fifo_mic, (void**)pFrame_mic->data, frame_mic_min_size);
                LeaveCriticalSection(&_section_mic);

                pFrame_spk->pts = av_frame_get_best_effort_timestamp(pFrame_spk);
                pFrame_mic->pts = av_frame_get_best_effort_timestamp(pFrame_mic);

                BufferSourceContext* s = (BufferSourceContext*)_filter_ctx_src_spk->priv;
                bool b1 = (s->sample_fmt != pFrame_spk->format);
                bool b2 = (s->sample_rate != pFrame_spk->sample_rate);
                bool b3 = (s->channel_layout != pFrame_spk->channel_layout);
                bool b4 = (s->channels != pFrame_spk->channels);

                ret = av_buffersrc_add_frame(_filter_ctx_src_spk, pFrame_spk);
                if (ret < 0)
                {
                    printf("Mixer: failed to call av_buffersrc_add_frame (speaker)\n");
                    break;
                }

                ret = av_buffersrc_add_frame(_filter_ctx_src_mic, pFrame_mic);
                if (ret < 0)
                {
                    printf("Mixer: failed to call av_buffersrc_add_frame (microphone)\n");
                    break;
                }

                while (1)
                {
                    AVFrame* pFrame_out = av_frame_alloc();

                    ret = av_buffersink_get_frame_flags(_filter_ctx_sink, pFrame_out, 0);
                    if (ret < 0)
                    {
                        printf("Mixer: failed to call av_buffersink_get_frame_flags\n");
                        break;
                    }
                    if (pFrame_out->data[0] != NULL)
                    {
                        av_init_packet(&packet_out);
                        packet_out.data = NULL;
                        packet_out.size = 0;

                        ret = avcodec_encode_audio2(_fmt_ctx_out->streams[_index_a_out]->codec, &packet_out, pFrame_out, &got_packet_ptr);
                        if (ret < 0)
                        {
                            printf("Mixer: failed to call avcodec_decode_audio4\n");
                            break;
                        }
                        if (got_packet_ptr)
                        {
                            packet_out.stream_index = _index_a_out;
                            packet_out.pts = frame_count * _fmt_ctx_out->streams[_index_a_out]->codec->frame_size;
                            packet_out.dts = packet_out.pts;
                            packet_out.duration = _fmt_ctx_out->streams[_index_a_out]->codec->frame_size;

                            packet_out.pts = av_rescale_q_rnd(packet_out.pts, 
                                _fmt_ctx_out->streams[_index_a_out]->codec->time_base,
                                _fmt_ctx_out->streams[_index_a_out]->time_base,
                                (AVRounding)(AV_ROUND_NEAR_INF | AV_ROUND_PASS_MINMAX));
                            packet_out.dts = packet_out.pts;
                            packet_out.duration = av_rescale_q_rnd(packet_out.duration,
                                _fmt_ctx_out->streams[_index_a_out]->codec->time_base,
                                _fmt_ctx_out->streams[_index_a_out]->time_base,
                                (AVRounding)(AV_ROUND_NEAR_INF | AV_ROUND_PASS_MINMAX));

                            frame_count++;

                            ret = av_interleaved_write_frame(_fmt_ctx_out, &packet_out);
                            if (ret < 0)
                            {
                                printf("Mixer: failed to call av_interleaved_write_frame\n");
                            }
                            printf("Mixer: write frame to file\n");
                        }
                        av_free_packet(&packet_out);                    
                    }
                    av_frame_free(&pFrame_out);
                }
            }
            else
            {
                tmpFifoFailed++;
                Sleep(20);
                if (tmpFifoFailed > 300)
                {
                    _state = CaptureState::STOPPED;
                    Sleep(30);
                    break;
                }
            }
            av_frame_free(&pFrame_spk);
            av_frame_free(&pFrame_mic);
        }
    }

    av_write_trailer(_fmt_ctx_out);

Release:
    av_audio_fifo_free(_fifo_spk);
    av_audio_fifo_free(_fifo_mic);

    avfilter_free(_filter_ctx_src_spk);
    avfilter_free(_filter_ctx_src_mic);
    avfilter_free(_filter_ctx_sink);

    avfilter_graph_free(&_filter_graph);


    if (_fmt_ctx_out)
    {
        avio_close(_fmt_ctx_out->pb);
    }

    avformat_close_input(&_fmt_ctx_spk);
    avformat_close_input(&_fmt_ctx_mic);
    avformat_free_context(_fmt_ctx_out);

    return ret;
}

下载地址

http://download.csdn.net/detail/dancing_night/9676014

dancing_night
关注
  • 0
    点赞
  • 踩
  • 14
    收藏
    觉得还不错? 一键收藏
  • 打赏
    打赏
  • 1
    评论
专栏目录
基于FFmpeg源码实现一个简单的混音demo，把文件a和文件b的音频混为一个音频流输出并存为文件
Fuguir™的博客
08-24 490
/* *一个简单的混音demo，把文件a和文件b的音频混为一个音频流输出并存为文件，只处理每个文件的第一个音频流 */ 命令行实现： ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT inputs The number of inputs. If unspecified, it defaults to 2.//输入...
(二) ffmpeg filter学习--混音实现
小小的熊的博客
08-05 3106
Audio 混音实现 从FFMPEG原生代码doc/examples/filtering_audio.c修改而来。 ffmpeg版本信息 1 2 3 4 5 6 7 8 9 10 11 ffmpeg version N-82997-g557c0df Copyright (c) ...
FFmpeg 混音学习【一】FFmpeg aac音频混流学习及问题记录
HiddenTrack的博客
07-11 1563
主要学习的是ffmpeg官方demo里的doc/examples/filtering_audio.c ffmpeg混流的主要流程 初始化filter 读取aac数据，解码后将数据压入filter buffer中 从buffer中取出数据编码成aac数据 1、初始化filter 首先定义需要使用到的变量： /** An instance of a filter */ static const char *filter_desc = "aresample=48000,aformat=sample
利用ffmpeg对两个音频文件进行混音处理
liuzhen007的专栏
03-11 603
原本程序自带的音频测试素材是一段时长 24 秒的 wav 文件，其中包含了关键词 Yes 和 No，两个关键词出现的具体时间点如下：想引入的噪点数据是前段时间在抖音上比较火的背景音乐——《野花与栀子花》，具体内容这里就省略了。
ffmpeg4教程10:avfilter混音处理
Java_lilin的专栏
12-21 1338
基于vs2017 vc++  ffmpeg4.0.2下测试 ffmpeg 环境配置请百度（vs2017 ffmpeg ）  主要是先将音频avframe采集到队列或链表里，再通过 av_buffersrc_add_frame添加多帧取出就是混音的贴出filter初始化的 字符串[in0][in1]amix=inputs=2[out]  其他的像写入空声音啥的请进群交流(讨论群261074...
FFMpeg-12、自带filter实现混音(系统音和麦克风音混合)
zw1996的博客
12-04 1828
获取系统音设备名称Setup Screen Capturer Recorder ，使用ffmpeg自带filter实现系统音和麦克风录制声音进行混音，以及一路静音的场景。
FFmpeg 混音学习【二】FFmpeg aac音频多路混流普适情况filter初始化
HiddenTrack的博客
07-13 797
前言 接上一篇博客，上一篇中介绍了单路和双路混流，要是输入输出流多了像之前那样写就会很麻烦，于是考虑普适的情况，输入一个n初始化好对应的filter。 具体实现 定义要用到的结构体和类(里面一些参数的含义可以看我上一篇博客) 输入流编码信息结构体： struct codec_info { int sample_rate; int channels; int64_t channel_layout; enum AVSampleFormat sample_fmt; AVRation
使用ffmpeg AVfilter 中的amix实现混音
好记性不如写博客！
08-03 9203
官方文档： 6.16 amix Mixes multiple audio inputs into a single output. Note that this filter only supports float samples (the amerge and pan audio filters support many formats). If the amix input has in
ffmpeg 音频混流--命令与代码说明
洛的博客
05-28 3521
  ffmpeg可使用filter实现各种复杂的操作，混音只是其中一小部分功能。代码实现因为参考已经说的较为详细，我这里补充一些我认为重要的。 1、命令实现 ffmpeg -i dragen.opus -i Monsters.opus -filter_complex "[0:a]aresample=8000[a0]; [1:a]adelay=delays=6400|6400,aresample=8000[a1];[a0][a1]amix=inputs=2" -b:a 20k -y mixer.aac -v
【Android FFMPEG 开发】Android 中使用 FFMPEG 进行混音操作
06-03
【Android FFMPEG 开发】Android 中使用 FFMPEG 进行混音操作 https://hanshuliang.blog.csdn.net/article/details/117526698 博客源码
ffmpeg 滤镜实现不同采样率多音频混音
03-17
利用ffmpeg开源库 对AAC数据进行解码，然后使用FFMPEG滤镜类对多个解码的PCM数据进行混合，最后对混合后的数据进行AAC编码； 主要内容： 音频解码，滤镜混音，音频编码 内容亮点： 当一个文件读取完毕后，程序不会...
ffmpeg 异步推理filter源码
03-21
本文旨在阐述如何通过FFmpeg开发一个创新的滤镜模块，该模块利用gRPC异步通信机制调用远程视频处理服务。这一设计实现了对ffmpeg滤镜功能的精简化整合，仅需通过配置ffmpeg命令行参数，即可实现不同算法间的灵活切换...
C# 利用ffmpeg进行推流
04-12
C# 利用ffmpeg进行推流，推流方式涉及：读取本地文件，以入读取指定内存数据；示例项目中使用开源项目 FFmpeg.AutoGen，其对ffmpeg的API做了封装，方便在C#代码中进行ffmpeg API的调用。
ffmpeg实现录屏+录音
热门推荐
dancing_night的专栏
06-30 3万+
1、概述 最简单的基于FFmpeg的AVDevice例子（屏幕录制+声音采集），视频采用mpeg4编码，音频采用aac编码，并生成mp4文件，其中fifo是此程序的关键，此程序只是一个demo很多优化都没做，仅供参考。 2、代码 /** *最简单的基于FFmpeg的AVDevice例子（屏幕录制+声音采集），视频采用mpeg4编码，音频采用aac编码，并生成mp4文件 *缪国凯 MK
ffmpeg源码跟踪笔记之av_write_frame 与 av_interleaved_write_frame
dancing_night的专栏
06-12 2万+
1、函数介绍 2、函数调用图 3、han
ffmpeg一些filter用法、以及一些功能命令
dancing_night的专栏
07-06 1万+
1、加字幕 命令：ffmpeg -i -filter_complex subtitles=filename=-y 说明：利用libass来为视频嵌入字幕，字幕是直接嵌入到视频里的硬字幕。 参考资料：http://ffmpeg.org/ffmpeg-filters.html#subtitles-1 2、剪切 命令：
ffmpeg调用libx264编码高级选项
dancing_night的专栏
04-23 1万+
在命令行中输入： ffmpeg -h -encoder=libx264 会输出一下信息： Encoder libx264 [libx264 H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10]: Threading capabilities: no Supported pixel formats: yuv420p yuvj420p yuv422
ffmpeg api 混音
最新发布
04-23
FFmpeg是一个开源的跨平台音视频处理工具，它提供了丰富的API用于音视频的编解码、转码、混音等操作。下面是关于FFmpeg API混音的介绍： FFmpeg提供了libavfilter库，其中包含了用于音频处理的滤镜。混音就是通过将...

“相关推荐”对你有帮助么？

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论 1
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包

打赏作者

dancing_night

你的鼓励将是我创作的最大动力

¥1 ¥2 ¥4 ¥6 ¥10 ¥20
扫码支付：¥1
获取中
扫码支付

您的余额不足，请更换扫码支付或充值

打赏作者

实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值