jmuxer

jmuxer

轻量级JavaScript MP4混合器:跨平台音视频处理解决方案

jMuxer是一款轻量级JavaScript MP4混合器,为浏览器和Node.js环境提供音视频处理功能。它支持H264视频和AAC音频输入,利用媒体源扩展实现浏览器播放,同时支持Node.js环境下的MP4导出。jMuxer采用协议无关设计,提供灵活的音视频处理选项,适用于多种流媒体应用场景。

jMuxerMP4复用器JavaScript媒体源扩展音视频处理Github开源项目

Build Status Maintenance license

jMuxer

jMuxer - a simple javascript mp4 muxer that works in both browser and node environment. It is communication protocol agnostic and it is intended to play media files on the browser with the help of the media source extension. It also can export mp4 on the node environment. It expects raw H264 video data and/or AAC audio data in ADTS container as an input.

Live Demo

Click here to view a working demo

Click here to play a h264 file online

How to use?

A distribution version is available on dist folder.

<script type="text/javascript" src="dist/jmuxer.min.js"></script> var jmuxer = new JMuxer(option);

Available options are:

node - String ID of a video tag / Reference of the HTMLVideoElement. Required field for browsers.

mode - Available values are: both, video and audio. Default is both

flushingTime - Buffer flushing time in milliseconds. Default value is 500 milliseconds. Set flushingTime to 0 if you want to flash buffer immediately or find any lag.

maxDelay - Maximum delay time in milliseconds. Default value is 500 milliseconds.

clearBuffer - true/false. Either it will clear played media buffer automatically or not. Default is true.

fps - Optional value. Frame rate of the video if it is known/fixed value. It will be used to find frame duration if chunk duration is not available with provided media data.

readFpsFromTrack - true/false. Will read FPS from MP4 track data instead of using (above) fps value. Default is false.

onReady - function. Will be called once MSE is ready.

onData - function. Will be called when muxed data is ready to be used. First argument is the muxed data.

onError - function. Will be fired if jMuxer encounters any buffer related errors.

onMissingVideoFrames - function. Will be fired if jMuxer encounters any missing video frames.

onMissingAudioFrames - function. Will be fired if jMuxer encounters any missing audio frames.

debug - true/false. Will print debug log in browser console. Default is false.

Complete example:

<script type="text/javascript" src="dist/jmuxer.min.js"></script> <video id="player"></video> <script> var jmuxer = new JMuxer({ node: 'player', mode: 'both', /* available values are: both, audio and video */ debug: false }); /* Now feed media data using feed method. audio and video is buffer data and duration is in milliseconds */ jmuxer.feed({ audio: audio, video: video, duration: duration }); </script>

Media dataObject may have following properties:

video - h264 buffer

audio - AAC buffer

duration - duration in milliseconds of the provided chunk. If duration is not provided, it will calculate frame duration wtih the provided frame rate (fps).

compositionTimeOffset - Composition time offset, difference between decode time and presentation time of frames, in milliseconds. This is only used for video and usually needed when B-frames are present in video stream.

ES6 Example:

Install module through npm

npm install --save jmuxer
import JMuxer from 'jmuxer'; const jmuxer = new JMuxer({ node: 'player', debug: true }); /* Now feed media data using feed method. audio and video is buffer data and duration is in milliseconds */ jmuxer.feed({ audio: audio, video: video, duration: duration });

Node Example:

Install module through npm

npm install --save jmuxer
const JMuxer = require('jmuxer'); const jmuxer = new JMuxer({ debug: true }); /* Stream in Object mode. Please check the example file for more details */ let h264_feeder = getFeederStreamSomehow(); let http_or_ws_or_any = getWritterStreamSomehow(); h264_feeder.pipe(jmuxer.createStream()).pipe(http_or_ws_or_any); // OR another way const jmuxer = new JMuxer({ onData: function(data) { res.write(data); // send data to client } debug: true }); jmuxer.feed({ audio: audio, video: video, duration: duration }); // feed data

Available Methods

NameParameterRemark
feeddata objectobject properites may have audio, video and duration. At least one media property i.e audio or video must be provided. If no duration is provided, it will calculate duration based on fps value
createStream-Get a writeable stream to feed buffer. Available on NodeJS only
reset-Reset the jmuxer and start over
destroy-Destroy the jmuxer instance and release the resources

Typescript definition

npm install --save @types/jmuxer

Compatibility

compatible with browsers supporting MSE with 'video/MP4. it is supported on:

  • Chrome for Android 34+
  • Chrome for Desktop 34+
  • Firefox for Android 41+
  • Firefox for Desktop 42+
  • IE11+ for Windows 8.1+
  • Edge for Windows 10+
  • Opera for Desktop
  • Safari for Mac 8+

Demo Server and player example

A simple node server and some demo media data are available in the example directory. In the example, each chunk/packet is consist of 4 bytes of header and the payload following the header. The first two bytes of the header contain the chunk duration and remaining two bytes contain the audio data length. Packet format is shown in the image below:

Packet format

2 bytes2 bytes
Duration (ms)Audio Data LengthAudio Data (AAC)Video Data (H264)

A step guideline to obtain above the packet format from your mp4 file using ffmpeg:

  1. Spliting video into 2 seconds chunks: ffmpeg -i input.mp4 -c copy -map 0 -segment_time 2 -f segment %03d.mp4
  2. Extracting h264 for all chunks: for f in *.mp4; do ffmpeg -i "$f" -vcodec copy -an -bsf:v h264_mp4toannexb "${f:0:3}.h264"; done
  3. Extracting audio for all chunks: for f in *.mp4; do ffmpeg -i "$f" -acodec copy -vn "${f:0:3}.aac"; done
  4. Extracting duration for all chunks: for f in *.mp4; do ffprobe "$f" -show_format 2>&1 | sed -n 's/duration=//p'; done

(see https://github.com/samirkumardas/jmuxer/issues/20#issuecomment-470855007)

How to run example?

Demo files are available in example directory. For running the example, first run the node server by following command:

cd example

node server.js

then, visit example/index.html page using any webserver.

Player Example for raw h264 only

Assuming you are still in example directory. Now run followngs:

node h264.js

then, visit example/h264.html page using any webserver.

How to build?

A distribution version is available inside dist directory. However, if you need to build, you can do as follows:

  1. git clone https://github.com/samirkumardas/jmuxer.git
  2. cd jmuxer
  3. npm install
  4. npm run build OR npm run pro

Support

If the project helps you, buy me a cup of coffee!

Credits

Proudly inspired by hls.js, rtsp player

Cobrowse.io - for sponsoring the adaptation of jMuxer for Node.js

编辑推荐精选

Trae

Trae

字节跳动发布的AI编程神器IDE

Trae是一种自适应的集成开发环境(IDE),通过自动化和多元协作改变开发流程。利用Trae,团队能够更快速、精确地编写和部署代码,从而提高编程效率和项目交付速度。Trae具备上下文感知和代码自动完成功能,是提升开发效率的理想工具。

AI工具TraeAI IDE协作生产力转型热门
问小白

问小白

全能AI智能助手,随时解答生活与工作的多样问题

问小白,由元石科技研发的AI智能助手,快速准确地解答各种生活和工作问题,包括但不限于搜索、规划和社交互动,帮助用户在日常生活中提高效率,轻松管理个人事务。

热门AI助手AI对话AI工具聊天机器人
Transly

Transly

实时语音翻译/同声传译工具

Transly是一个多场景的AI大语言模型驱动的同声传译、专业翻译助手,它拥有超精准的音频识别翻译能力,几乎零延迟的使用体验和支持多国语言可以让你带它走遍全球,无论你是留学生、商务人士、韩剧美剧爱好者,还是出国游玩、多国会议、跨国追星等等,都可以满足你所有需要同传的场景需求,线上线下通用,扫除语言障碍,让全世界的语言交流不再有国界。

讯飞智文

讯飞智文

一键生成PPT和Word,让学习生活更轻松

讯飞智文是一个利用 AI 技术的项目,能够帮助用户生成 PPT 以及各类文档。无论是商业领域的市场分析报告、年度目标制定,还是学生群体的职业生涯规划、实习避坑指南,亦或是活动策划、旅游攻略等内容,它都能提供支持,帮助用户精准表达,轻松呈现各种信息。

AI办公办公工具AI工具讯飞智文AI在线生成PPTAI撰写助手多语种文档生成AI自动配图热门
讯飞星火

讯飞星火

深度推理能力全新升级,全面对标OpenAI o1

科大讯飞的星火大模型,支持语言理解、知识问答和文本创作等多功能,适用于多种文件和业务场景,提升办公和日常生活的效率。讯飞星火是一个提供丰富智能服务的平台,涵盖科技资讯、图像创作、写作辅助、编程解答、科研文献解读等功能,能为不同需求的用户提供便捷高效的帮助,助力用户轻松获取信息、解决问题,满足多样化使用场景。

热门AI开发模型训练AI工具讯飞星火大模型智能问答内容创作多语种支持智慧生活
Spark-TTS

Spark-TTS

一种基于大语言模型的高效单流解耦语音令牌文本到语音合成模型

Spark-TTS 是一个基于 PyTorch 的开源文本到语音合成项目,由多个知名机构联合参与。该项目提供了高效的 LLM(大语言模型)驱动的语音合成方案,支持语音克隆和语音创建功能,可通过命令行界面(CLI)和 Web UI 两种方式使用。用户可以根据需求调整语音的性别、音高、速度等参数,生成高质量的语音。该项目适用于多种场景,如有声读物制作、智能语音助手开发等。

咔片PPT

咔片PPT

AI助力,做PPT更简单!

咔片是一款轻量化在线演示设计工具,借助 AI 技术,实现从内容生成到智能设计的一站式 PPT 制作服务。支持多种文档格式导入生成 PPT,提供海量模板、智能美化、素材替换等功能,适用于销售、教师、学生等各类人群,能高效制作出高品质 PPT,满足不同场景演示需求。

讯飞绘文

讯飞绘文

选题、配图、成文,一站式创作,让内容运营更高效

讯飞绘文,一个AI集成平台,支持写作、选题、配图、排版和发布。高效生成适用于各类媒体的定制内容,加速品牌传播,提升内容营销效果。

热门AI辅助写作AI工具讯飞绘文内容运营AI创作个性化文章多平台分发AI助手
材料星

材料星

专业的AI公文写作平台,公文写作神器

AI 材料星,专业的 AI 公文写作辅助平台,为体制内工作人员提供高效的公文写作解决方案。拥有海量公文文库、9 大核心 AI 功能,支持 30 + 文稿类型生成,助力快速完成领导讲话、工作总结、述职报告等材料,提升办公效率,是体制打工人的得力写作神器。

openai-agents-python

openai-agents-python

OpenAI Agents SDK,助力开发者便捷使用 OpenAI 相关功能。

openai-agents-python 是 OpenAI 推出的一款强大 Python SDK,它为开发者提供了与 OpenAI 模型交互的高效工具,支持工具调用、结果处理、追踪等功能,涵盖多种应用场景,如研究助手、财务研究等,能显著提升开发效率,让开发者更轻松地利用 OpenAI 的技术优势。

下拉加载更多