ffmpeg为mkv封装格式的音视频文件添加内挂字幕

现在好莱坞的电影，都是全球看，一个地区的人看电影时，电影屏幕上应该展示对应的本地区语言字幕。故电影画面在不同的地区，需要配置不同的语言字幕。故视频画面里面的字幕应该可以拆出来，不能像老版三国演义，每到经典处，展示出文字，如下所示：

这种文字是直接嵌入到视频画面，无法拆出来，这种字幕也叫内嵌字幕。

本文要讲的是内挂字幕，字幕在视频文件里面，但是是独立的通道，可以独立拆出来。当然，还有一种外挂字幕，是在视频文件外面，播放器播放时，可以选择本地的字幕文件。

就封装格式而言，目前mkv对字幕支持的最好，读者可以先准备下字幕文件，字幕文件，读者可以网上下载现有的，也可以自己制作，本文准备的字幕文件ts.ass的内容如下：

[Script Info]
Title: Untitled
ScriptType: v4.00+
PlayResX:1280
PlayResY:720
WrapStyle: 0
ScaledBorderAndShadow: yes[V4+ Styles]
Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
Style: Default,Arial,20,&H00FFFFFF,&H000000FF,&H00000000,&H00000000,0,0,0,0,100,100,0,0,1,2,2,2,10,10,10,1[Events]
Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
Dialogue: 2,0:00:00.22,0:00:31.93,Default,,0,0,0,,就算身处 流逝的时光里
Dialogue: 0,0:00:32.02,0:00:36.18,Default,,0,0,0,,也只有倦怠 在原地打转不停
Dialogue: 0,0:00:36.19,0:00:38.85,Default,,0,0,0,,从我身边 渐行渐远的心
Dialogue: 0,0:00:39.03,0:00:43.15,Default,,0,0,0,,再也模糊不清 你明白吗
Dialogue: 0,0:00:43.20,0:00:45.67,Default,,0,0,0,,我的身体 已经动弹不得
Dialogue: 0,0:00:45.79,0:00:50.16,Default,,0,0,0,,在时间的狭缝里 随波逐流
Dialogue: 0,0:00:50.17,0:00:53.26,Default,,0,0,0,,周围的一切 都与我无关
Dialogue: 0,0:00:53.39,0:00:57.07,Default,,0,0,0,,我就是我 仅·此·而·已
Dialogue: 0,0:00:57.18,0:01:00.04,Default,,0,0,0,,我在做梦吗？什么都没在看
Dialogue: 0,0:01:00.18,0:01:03.56,Default,,0,0,0,,出口也是枉然 自怜自艾的废话
Dialogue: 0,0:01:03.66,0:01:07.08,Default,,0,0,0,,悲伤什么的 只会徒增疲倦啊
Dialogue: 0,0:01:07.21,0:01:10.52,Default,,0,0,0,,干脆就这样 在麻木中度日吧
Dialogue: 0,0:01:10.62,0:01:13.95,Default,,0,0,0,,就算被灌以 喧嚣的闲言碎语
Dialogue: 0,0:01:14.09,0:01:17.45,Default,,0,0,0,,我的心也已经 不再起一丝涟漪
Dialogue: 0,0:01:17.56,0:01:21.03,Default,,0,0,0,,如果我能够 驱使自己的话
Dialogue: 0,0:01:21.13,0:01:24.39,Default,,0,0,0,,就让这一切 被黑暗所吞没吧
Dialogue: 0,0:01:24.50,0:01:28.06,Default,,0,0,0,,这样的我 还有未来可言吗
Dialogue: 0,0:01:28.19,0:01:31.46,Default,,0,0,0,,这种世界 允许我的存在吗
Dialogue: 0,0:01:31.56,0:01:34.85,Default,,0,0,0,,此刻感到窒息吗？此刻觉得悲伤吗
Dialogue: 0,0:01:34.98,0:01:38.45,Default,,0,0,0,,就连自己的事 也根本搞不懂啊
Dialogue: 0,0:01:38.55,0:01:41.94,Default,,0,0,0,,就算走下去 也只是徒增疲倦
Dialogue: 0,0:01:42.06,0:01:45.32,Default,,0,0,0,,对他人的一切 完全无法理解
Dialogue: 0,0:01:45.42,0:01:48.64,Default,,0,0,0,,这样的我 如果还能改变
Dialogue: 0,0:01:48.78,0:01:52.14,Default,,0,0,0,,还能改变的话 可以化为空白吗
Dialogue: 0,0:02:06.79,0:02:09.33,Default,,0,0,0,,就算身处 流逝的时光里
Dialogue: 0,0:02:09.48,0:02:13.64,Default,,0,0,0,,也只有倦怠 在原地打转不停
Dialogue: 0,0:02:13.77,0:02:16.24,Default,,0,0,0,,从我身边 渐行渐远的心
Dialogue: 0,0:02:16.37,0:02:20.54,Default,,0,0,0,,再也模糊不清 你明白吗
Dialogue: 0,0:02:20.67,0:02:23.26,Default,,0,0,0,,我的身体 已经动弹不得
Dialogue: 0,0:02:23.36,0:02:27.67,Default,,0,0,0,,在时间的狭缝里 随波逐流
Dialogue: 0,0:02:27.68,0:02:31.19,Default,,0,0,0,,周围的一切 都与我无关
Dialogue: 0,0:02:31.28,0:02:34.31,Default,,0,0,0,,我就是我 仅·此·而·已
Dialogue: 0,0:02:34.49,0:02:37.50,Default,,0,0,0,,我在做梦吗？什么都没在看
Dialogue: 0,0:02:37.62,0:02:40.93,Default,,0,0,0,,出口也是枉然 自怜自艾的废话
Dialogue: 0,0:02:41.03,0:02:44.30,Default,,0,0,0,,悲伤什么的 只会徒增疲倦啊
Dialogue: 0,0:02:44.43,0:02:47.91,Default,,0,0,0,,干脆就这样 在麻木中度日吧
Dialogue: 0,0:02:47.99,0:02:51.61,Default,,0,0,0,,就算被灌以 喧嚣的闲言碎语
Dialogue: 0,0:02:51.72,0:02:54.73,Default,,0,0,0,,我的心也已经 不再起一丝涟漪
Dialogue: 0,0:02:54.82,0:02:58.30,Default,,0,0,0,,如果我能够 驱使自己的话
Dialogue: 0,0:02:58.39,0:03:02.04,Default,,0,0,0,,就让这一切 被黑暗所吞没吧
Dialogue: 0,0:03:02.05,0:03:05.39,Default,,0,0,0,,如果任我驱使 驱使自己的话
Dialogue: 0,0:03:05.47,0:03:08.92,Default,,0,0,0,,一切都会毁灭 一切都会毁灭啊
Dialogue: 0,0:03:09.03,0:03:12.36,Default,,0,0,0,,被悲伤笼罩 被悲伤笼罩的话
Dialogue: 0,0:03:12.47,0:03:15.74,Default,,0,0,0,,我的心还能够 化为空白吗
Dialogue: 0,0:03:15.85,0:03:19.25,Default,,0,0,0,,不论你的存在 还是我的存在
Dialogue: 0,0:03:19.35,0:03:22.67,Default,,0,0,0,,这一切的真实 我都一无所知
Dialogue: 0,0:03:22.79,0:03:26.18,Default,,0,0,0,,如果在此睁开 这沉重的双眼
Dialogue: 0,0:03:26.36,0:03:29.88,Default,,0,0,0,,一切都会毁灭 被黑暗所吞没

读者可以清晰的看到，哪段时间至哪段时间，界面需要展示的文字，比如最后一个Dialogue显示在03:26.36到0:03:29.88这段时间，界面应该展示一切都会毁灭被黑暗所吞没。

可以通过如下的ffmpeg命令降此字幕内挂到视频文件中

ffmpeg -i TAEYEON-Weekend.mkv -i ts.ass -c copy output.mkv

下面用ffmpeg代码的方式展示如何实现。
首先，需要说明的是，字幕跟音频，视频一样，有自己的通道，有自己的time_base，其读取方法也是av_read_frame。这点跟内嵌字幕不一样，在一个视频中，添加内嵌文字，可以通过滤镜drawtext实现，有解码，滤镜运算，编码过程，很费时，内挂不一样，没有这三个费时的计算，故往视频文件中添加内挂字幕很快。

其次，本人通过两个队列m_vecMediaPacket和m_vecAssPacket来存储读取的packet，然后在一个线程里面按照写入时间顺序分别写入m_vecMediaPacket和m_vecAssPacket的数据。

std::deque<AVPacket *> m_vecMediaPacket;
std::deque<AVPacket *> m_vecAssPacket;

此处，本人在av_read_frame，得到AVPacket后，没有直接调用av_interleaved_write_frame写文件，最主要的原因是av_interleaved_write_frame里面会对AVPacket的时间(相对各自的AVStream)进行排序，若视频文件比较大，则可能里面需要分配的空间也越来越大，最终由于内存不足导致崩溃。
故本人将读取的音视频packet和字幕packet分别存入队列，然后按照音视频播放同步的原理，调用av_write_frame依次写入m_vecMediaPacket和m_vecAssPacket里面的内容。也就是音视频的packet和字幕的packet，由自己编码判断谁先写(代码中av_compare_ts部分)，而不是交由av_interleaved_write_frame处理。

再次，本人讲解下代码的大致结构：
1.用avformat_open_input分别打开媒体文件和字幕文件
2.avformat_alloc_output_context2构建输出文件context后，用avformat_new_stream分别往里面添加媒体流和字幕流，代码如下：

int iStreamNum = m_pFormatCtx_MediaFile->nb_streams;
for (int i = 0; i < iStreamNum; i++)
{AVCodec* pCodecEncode_Media = (AVCodec *)avcodec_find_encoder(m_pFormatCtx_MediaFile->streams[i]->codecpar->codec_id);AVStream *pMediaStream = avformat_new_stream(m_pFormatCtx_Out, pCodecEncode_Media);if (!pCodecEncode_Media){break;}avcodec_parameters_copy(pMediaStream->codecpar, m_pFormatCtx_MediaFile->streams[i]->codecpar);pMediaStream->codecpar->codec_tag = 0;
}{AVCodec* pCodecEncode_Ass = (AVCodec *)avcodec_find_encoder(m_pFormatCtx_AssFile->streams[0]->codecpar->codec_id);AVStream *pAssStream = avformat_new_stream(m_pFormatCtx_Out, pCodecEncode_Ass);if (!pAssStream){break;}avcodec_parameters_copy(pAssStream->codecpar, m_pFormatCtx_AssFile->streams[0]->codecpar);pAssStream->codecpar->codec_tag = 0;
}

3.创建三个线程，如下所示：

m_hMediaFileReadThread = CreateThread(NULL, 0, MediaFileReadProc, this, 0, NULL);
m_hAssFileReadThread = CreateThread(NULL, 0, AssFileReadProc, this, 0, NULL);
m_hWriteThread = CreateThread(NULL, 0, WriteProc, this, 0, NULL);

前两个线程分别读取媒体流和字幕流，然后塞入队列，第三个线程读取两个队列中的数据，然后按照时间顺序写入packet。
这里说明下，对于字幕流而言，av_read_frame，本人调用到了48次，这48次，其实也是上面的ass文件中，Dialogue节点的数量。

最后，是代码，文件结构如下：

其中FfmpegMkvTest.cpp内容如下：

#include <iostream>
#include "FfmpegAddAss.h"int main()
{CFfmpegAddAss cFfmpegAddAss;std::string strMediaFile = "D:/learn/ffmpeg/FfmpegConvert/x64/Release/TAEYEON-Weekend.mkv";std::string strAssFile = "D:/learn/ffmpeg/FfmpegConvert/x64/Release/ts.ass";std::string strOutFile = "D:/learn/ffmpeg/FfmpegConvert/x64/Release/TAEYEON-Weekend_ass.mkv";cFfmpegAddAss.StartAddAss(strMediaFile, strAssFile, strOutFile);cFfmpegAddAss.WaitFinish();return 0;
}

其中WaitFinish函数在文件处理结束后，会返回。

FfmpegAddAss.h的内容如下：

#pragma once#include <string>
#include <Windows.h>
#include <deque>#define MAX_PACKET_NUM 200#ifdef  __cplusplus
extern "C"
{#endif
#include "libavcodec/avcodec.h"
#include "libavformat/avformat.h"
#include "libswscale/swscale.h"
#include "libswresample/swresample.h"
#include "libavdevice/avdevice.h"
#include "libavutil/audio_fifo.h"
#include "libavutil/avutil.h"
#include "libavutil/fifo.h"
#include "libavutil/frame.h"
#include "libavutil/imgutils.h"#include "libavfilter/avfilter.h"
#include "libavfilter/buffersink.h"
#include "libavfilter/buffersrc.h"#ifdef __cplusplus
};
#endifclass CFfmpegAddAss
{public:CFfmpegAddAss();~CFfmpegAddAss();
public:int StartAddAss(std::string strMediaFile, std::string strAssFile, std::string strOutFile);void WaitFinish();
private:int OpenMediaFile(std::string strMediaFile);int OpenAssFile(std::string strAssFile);int OpenOutFile(std::string strOutFile);
private:static DWORD WINAPI MediaFileReadProc(LPVOID lpParam);void MediaFileRead();static DWORD WINAPI AssFileReadProc(LPVOID lpParam);void AssFileRead();static DWORD WINAPI WriteProc(LPVOID lpParam);void Write();
private:AVFormatContext *m_pFormatCtx_MediaFile = NULL;AVFormatContext *m_pFormatCtx_AssFile = NULL;AVFormatContext *m_pFormatCtx_Out = NULL;int m_iAssStreamIndex = -1;HANDLE m_hMediaFileReadThread = NULL;HANDLE m_hAssFileReadThread = NULL;HANDLE m_hWriteThread = NULL;std::deque<AVPacket *> m_vecMediaPacket;std::deque<AVPacket *> m_vecAssPacket;CRITICAL_SECTION m_csMediaSection;CRITICAL_SECTION m_csAssSection;bool m_bStart = false;
};

FfmpegAddAss.cpp内容如下：

#include "FfmpegAddAss.h"#ifdef    __cplusplus
extern "C"
{#endif#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, "postproc.lib")
#pragma comment(lib, "swresample.lib")
#pragma comment(lib, "swscale.lib")#ifdef __cplusplus
};
#endifCFfmpegAddAss::CFfmpegAddAss()
{InitializeCriticalSection(&m_csMediaSection);InitializeCriticalSection(&m_csAssSection);
}CFfmpegAddAss::~CFfmpegAddAss()
{DeleteCriticalSection(&m_csMediaSection);DeleteCriticalSection(&m_csAssSection);
}int CFfmpegAddAss::StartAddAss(std::string strMediaFile, std::string strAssFile, std::string strOutFile)
{int ret = 0;do{ret = OpenMediaFile(strMediaFile);if (ret < 0){break;}ret = OpenAssFile(strAssFile);if (ret < 0){break;}ret = OpenOutFile(strOutFile);if (ret < 0){break;}m_bStart = true;m_hMediaFileReadThread = CreateThread(NULL, 0, MediaFileReadProc, this, 0, NULL);m_hAssFileReadThread = CreateThread(NULL, 0, AssFileReadProc, this, 0, NULL);m_hWriteThread = CreateThread(NULL, 0, WriteProc, this, 0, NULL);} while (0);return ret;
}void CFfmpegAddAss::WaitFinish()
{DWORD dw = 0;for (int i = 0; i < 10000; i++){if (m_hMediaFileReadThread == NULL && m_hAssFileReadThread == NULL){break;}if (m_hMediaFileReadThread != NULL){dw = WaitForSingleObject(m_hMediaFileReadThread, 1000);if (dw == WAIT_OBJECT_0){CloseHandle(m_hMediaFileReadThread);m_hMediaFileReadThread = NULL;}}if (m_hAssFileReadThread != NULL){dw = WaitForSingleObject(m_hAssFileReadThread, 1000);if (dw == WAIT_OBJECT_0){CloseHandle(m_hAssFileReadThread);m_hAssFileReadThread = NULL;}}}while (m_vecMediaPacket.size() > 0 && m_vecAssPacket.size() > 0){Sleep(1000);}Sleep(1000);m_bStart = false;WaitForSingleObject(m_hWriteThread, INFINITE);CloseHandle(m_hWriteThread);m_hWriteThread = NULL;
}int CFfmpegAddAss::OpenMediaFile(std::string strMediaFile)
{int ret = -1;do{if ((ret = avformat_open_input(&m_pFormatCtx_MediaFile, strMediaFile.c_str(), 0, 0)) < 0) {break;}if ((ret = avformat_find_stream_info(m_pFormatCtx_MediaFile, 0)) < 0) {break;}ret = 0;} while (0);return ret;
}int CFfmpegAddAss::OpenAssFile(std::string strAssFile)
{int ret = -1;do{if ((ret = avformat_open_input(&m_pFormatCtx_AssFile, strAssFile.c_str(), 0, 0)) < 0) {break;}if ((ret = avformat_find_stream_info(m_pFormatCtx_AssFile, 0)) < 0) {break;}ret = 0;} while (0);return ret;
}int CFfmpegAddAss::OpenOutFile(std::string strOutFile)
{int iRet = -1;do{avformat_alloc_output_context2(&m_pFormatCtx_Out, NULL, NULL, strOutFile.c_str());int iStreamNum = m_pFormatCtx_MediaFile->nb_streams;for (int i = 0; i < iStreamNum; i++){AVCodec* pCodecEncode_Media = (AVCodec *)avcodec_find_encoder(m_pFormatCtx_MediaFile->streams[i]->codecpar->codec_id);AVStream *pMediaStream = avformat_new_stream(m_pFormatCtx_Out, pCodecEncode_Media);if (!pCodecEncode_Media){break;}avcodec_parameters_copy(pMediaStream->codecpar, m_pFormatCtx_MediaFile->streams[i]->codecpar);pMediaStream->codecpar->codec_tag = 0;}{AVCodec* pCodecEncode_Ass = (AVCodec *)avcodec_find_encoder(m_pFormatCtx_AssFile->streams[0]->codecpar->codec_id);AVStream *pAssStream = avformat_new_stream(m_pFormatCtx_Out, pCodecEncode_Ass);if (!pAssStream){break;}avcodec_parameters_copy(pAssStream->codecpar, m_pFormatCtx_AssFile->streams[0]->codecpar);pAssStream->codecpar->codec_tag = 0;}if (!(m_pFormatCtx_Out->oformat->flags & AVFMT_NOFILE)){if (avio_open(&m_pFormatCtx_Out->pb, strOutFile.c_str(), AVIO_FLAG_WRITE) < 0){break;}}if (avformat_write_header(m_pFormatCtx_Out, NULL) < 0){break;}m_iAssStreamIndex = iStreamNum;iRet = 0;} while (0);if (iRet != 0){if (m_pFormatCtx_Out != NULL){avformat_free_context(m_pFormatCtx_Out);m_pFormatCtx_Out = NULL;}}return iRet;
}DWORD WINAPI CFfmpegAddAss::MediaFileReadProc(LPVOID lpParam)
{CFfmpegAddAss *pFfmpegAddAss = (CFfmpegAddAss *)lpParam;if (pFfmpegAddAss != NULL){pFfmpegAddAss->MediaFileRead();}return 0;
}void CFfmpegAddAss::MediaFileRead()
{AVPacket packet = { 0 };int ret = 0;while (1){av_packet_unref(&packet);ret = av_read_frame(m_pFormatCtx_MediaFile, &packet);if (ret == AVERROR(EAGAIN)){continue;}else if (ret == AVERROR_EOF){break;}else if (ret < 0){break;}while (1){int iPacketNum = m_vecMediaPacket.size();if (iPacketNum >= MAX_PACKET_NUM){Sleep(10);continue;}else{AVPacket *pPacket = av_packet_clone(&packet);if (pPacket != NULL){EnterCriticalSection(&m_csMediaSection);m_vecMediaPacket.push_back(pPacket);LeaveCriticalSection(&m_csMediaSection);}}break;}}
}DWORD WINAPI CFfmpegAddAss::AssFileReadProc(LPVOID lpParam)
{CFfmpegAddAss *pFfmpegAddAss = (CFfmpegAddAss *)lpParam;if (pFfmpegAddAss != NULL){pFfmpegAddAss->AssFileRead();}return 0;
}void CFfmpegAddAss::AssFileRead()
{AVPacket packet = { 0 };int ret = 0;while (1){av_packet_unref(&packet);ret = av_read_frame(m_pFormatCtx_AssFile, &packet);if (ret == AVERROR(EAGAIN)){continue;}else if (ret == AVERROR_EOF){break;}else if (ret < 0){break;}while (1){int iPacketNum = m_vecAssPacket.size();if (iPacketNum >= MAX_PACKET_NUM){Sleep(10);continue;}else{AVPacket *pPacket = av_packet_clone(&packet);if (pPacket != NULL){EnterCriticalSection(&m_csAssSection);m_vecAssPacket.push_back(pPacket);LeaveCriticalSection(&m_csAssSection);}}break;}}
}DWORD WINAPI CFfmpegAddAss::WriteProc(LPVOID lpParam)
{CFfmpegAddAss *pFfmpegAddAss = (CFfmpegAddAss *)lpParam;if (pFfmpegAddAss != NULL){pFfmpegAddAss->Write();}return 0;
}void CFfmpegAddAss::Write()
{int ret = 0;int cur_pts_media = 0;int cur_pts_ass = 0;AVPacket packet = { 0 };int iPicCount = 0;int iMediaIndex = 0;while (m_bStart){if (av_compare_ts(cur_pts_media, m_pFormatCtx_Out->streams[iMediaIndex]->time_base,cur_pts_ass, m_pFormatCtx_Out->streams[m_iAssStreamIndex]->time_base) <= 0){int iPacketNum = m_vecMediaPacket.size();if (iPacketNum >= 1){AVPacket *pPacket = NULL;EnterCriticalSection(&m_csMediaSection);if (!m_vecMediaPacket.empty()){pPacket = m_vecMediaPacket.front();m_vecMediaPacket.pop_front();}LeaveCriticalSection(&m_csMediaSection);pPacket->pts = av_rescale_q_rnd(pPacket->pts, m_pFormatCtx_MediaFile->streams[pPacket->stream_index]->time_base, m_pFormatCtx_Out->streams[m_iAssStreamIndex]->time_base, AVRounding(1));pPacket->dts = av_rescale_q_rnd(pPacket->dts, m_pFormatCtx_MediaFile->streams[pPacket->stream_index]->time_base, m_pFormatCtx_Out->streams[m_iAssStreamIndex]->time_base, AVRounding(1));pPacket->duration = av_rescale_q_rnd(pPacket->duration, m_pFormatCtx_MediaFile->streams[pPacket->stream_index]->time_base, m_pFormatCtx_Out->streams[m_iAssStreamIndex]->time_base, AVRounding(1));cur_pts_media = pPacket->pts;iMediaIndex = pPacket->stream_index;ret = av_write_frame(m_pFormatCtx_Out, pPacket);av_packet_free(&pPacket);}else{Sleep(1);if (m_hMediaFileReadThread == NULL){break;}}}else{int iPacketNum = m_vecAssPacket.size();if (iPacketNum >= 1){AVPacket *pPacket = NULL;EnterCriticalSection(&m_csAssSection);if (!m_vecAssPacket.empty()){pPacket = m_vecAssPacket.front();m_vecAssPacket.pop_front();}LeaveCriticalSection(&m_csAssSection);pPacket->pts = av_rescale_q_rnd(pPacket->pts, m_pFormatCtx_AssFile->streams[0]->time_base, m_pFormatCtx_Out->streams[m_iAssStreamIndex]->time_base, AVRounding(1));pPacket->dts = av_rescale_q_rnd(pPacket->dts, m_pFormatCtx_AssFile->streams[0]->time_base, m_pFormatCtx_Out->streams[m_iAssStreamIndex]->time_base, AVRounding(1));pPacket->duration = av_rescale_q_rnd(pPacket->duration, m_pFormatCtx_AssFile->streams[0]->time_base, m_pFormatCtx_Out->streams[m_iAssStreamIndex]->time_base, AVRounding(1));pPacket->stream_index = m_iAssStreamIndex;cur_pts_ass = pPacket->pts;ret = av_write_frame(m_pFormatCtx_Out, pPacket);av_packet_free(&pPacket);}else{Sleep(1);if (m_hAssFileReadThread == NULL){break;}}}}while (m_hMediaFileReadThread != NULL || m_vecMediaPacket.size() >= 1){AVPacket *pPacket = NULL;EnterCriticalSection(&m_csMediaSection);if (!m_vecMediaPacket.empty()){pPacket = m_vecMediaPacket.front();m_vecMediaPacket.pop_front();}else{LeaveCriticalSection(&m_csMediaSection);continue;}LeaveCriticalSection(&m_csMediaSection);pPacket->pts = av_rescale_q_rnd(pPacket->pts, m_pFormatCtx_MediaFile->streams[pPacket->stream_index]->time_base, m_pFormatCtx_Out->streams[m_iAssStreamIndex]->time_base, AVRounding(1));pPacket->dts = av_rescale_q_rnd(pPacket->dts, m_pFormatCtx_MediaFile->streams[pPacket->stream_index]->time_base, m_pFormatCtx_Out->streams[m_iAssStreamIndex]->time_base, AVRounding(1));pPacket->duration = av_rescale_q_rnd(pPacket->duration, m_pFormatCtx_MediaFile->streams[pPacket->stream_index]->time_base, m_pFormatCtx_Out->streams[m_iAssStreamIndex]->time_base, AVRounding(1));cur_pts_media = pPacket->pts;iMediaIndex = pPacket->stream_index;ret = av_write_frame(m_pFormatCtx_Out, pPacket);av_packet_free(&pPacket);}while (m_hAssFileReadThread != NULL || m_vecAssPacket.size() >= 1){AVPacket *pPacket = NULL;EnterCriticalSection(&m_csAssSection);if (!m_vecAssPacket.empty()){pPacket = m_vecAssPacket.front();m_vecAssPacket.pop_front();}else{LeaveCriticalSection(&m_csAssSection);continue;}LeaveCriticalSection(&m_csAssSection);pPacket->pts = av_rescale_q_rnd(pPacket->pts, m_pFormatCtx_AssFile->streams[0]->time_base, m_pFormatCtx_Out->streams[m_iAssStreamIndex]->time_base, AVRounding(1));pPacket->dts = av_rescale_q_rnd(pPacket->dts, m_pFormatCtx_AssFile->streams[0]->time_base, m_pFormatCtx_Out->streams[m_iAssStreamIndex]->time_base, AVRounding(1));pPacket->duration = av_rescale_q_rnd(pPacket->duration, m_pFormatCtx_AssFile->streams[0]->time_base, m_pFormatCtx_Out->streams[m_iAssStreamIndex]->time_base, AVRounding(1));pPacket->stream_index = m_iAssStreamIndex;cur_pts_ass = pPacket->pts;ret = av_write_frame(m_pFormatCtx_Out, pPacket);av_packet_free(&pPacket);}Sleep(100);av_write_trailer(m_pFormatCtx_Out);avio_close(m_pFormatCtx_Out->pb);
}

运行效果如下：

本视频4分钟7秒，而ass文件只有3分30多秒，故最后一段视频上，没字幕。
下面是字幕通道。

ffmpeg为mkv封装格式的音视频文件添加内挂字幕相关推荐

如何把视频文件添加中英文对照字幕
如何把视频文件添加中英文对照字幕在这里用到了subtitle edit的软件,首先我们先安装好subtitle和vlc软件,见上一篇文章下载Subtitle edit和vlc 1,点击视频,打开视频 ...
C++实现flv封装格式解析(音视频学习笔记三)
这篇博文使用C++解析一个flv文件信息,对其中一些重要的信息进行log输出,对flv的数据封装格式信息不清楚的可以去看这篇博文-FLV 封装格式解析,里面详细说明了flv文件的结构信息.这篇博文参考 ...
07-----给音视频文件添加字幕流
添加字幕流可以使用ACHTIME软件去添加. 一下载ACHTIME软件在下载之前我们需要注册一个账号.然后才能下载. http://arctime.cn/download.html 二添加字幕流 ...
音视频基础之封装格式与音视频同步
封装格式的概念封装格式(也叫容器)就是将已经编码压缩好的视频流.音频流及字幕按照一定的方案放到一个文件中,便于播放软件播放. 一般来说,视频文件的后缀名就是它的封装格式. 封装的格式不一样,后缀名也 ...
前端教程：浏览器不兼容嵌入的音视频文件怎么办？
虽然HTML5支持ogg.mpeg4和webm的视频格式以及ogg.mp3和wav的音频格式,但并不是所有的浏览器都支持这些格式,因此我们在嵌入视频音频文件格式时,就要考虑浏览器的兼容性问题.表1列举 ...
常见音视频文件的编码和封装格式详解
常见的AVI.RMVB.MKV.ASF.WMV.MP4.3GP.FLV等⽂件其实只能算是⼀种封装标准. ⼀个完整的视频⽂件是由⾳频和视频2部分组成的.H264.Xvid等就是视频编码格式,MP3.AA ...
ffmpeg音视频文件音视频流抽取，初步尝试人声分离
文章目录 ffmpeg抽取音视频文件中的音频流音频流类型 AAC与m4a的区别 AAC与mp3的区别用ffmpeg查看视频的信息用ffmpeg抽取AAC音频流从AAC文件中获取音轨音轨是什么 ...
ffmpeg学习（11）音视频文件muxer（2）多输入混流
在 ffmpeg学习(3)编码.解码的流程介绍和 ffmpeg学习(9)音视频文件demuxer中介绍了媒体文件的解封装.本文记录Ffmpeg封装格式另一种处理与与demuxer相反方式–视音频复用 ...
音视频文件的容器格式和编码格式
转自:http://blog.chinaunix.net/uid-25885064-id-3338166.html 音视频文件的容器格式和编码格式视频和音频以视频文件格式的形式放在一个文件中,方便同 ...

ffmpeg为mkv封装格式的音视频文件添加内挂字幕

ffmpeg为mkv封装格式的音视频文件添加内挂字幕相关推荐

最新文章

热门文章