Project Icon

hls-server

轻量级HTTP Live Streaming媒体流服务器中间件

hls-server是一个开源的Node.js HTTP Live Streaming (HLS)媒体流服务中间件。它支持文件系统和内存流两种方式,易于集成到现有HTTP服务器中。该项目提供CLI工具,便于快速部署HLS服务。支持FFMPEG编码和分段,适用于点播和直播场景,是搭建流媒体服务的实用工具。

hls-server

JavaScript Style Guide Travis

Simple HTTP middleware for serving HTTP Live Streaming (HLS) compatible media streams.

Usage

First you need a compatible media stream (see Producing Streams).

Fast way:

require('hls-server')(8000)

Detailed way:

var HLSServer = require('hls-server')
var http = require('http')

var server = http.createServer()
var hls = new HLSServer(server, {
  path: '/streams',     // Base URI to output HLS streams
  dir: 'public/videos'  // Directory that input files are stored
})
server.listen(8000)

Producing Streams

HLS can only stream files that have been properly encoded and segmented. FFMPEG is great for this.
Here is how to do it with node-fluent-ffmpeg.

var ffmpeg = require('fluent-ffmpeg')

function callback() { // do something when encoding is done }

// Below is FFMPEG converting MP4 to HLS with reasonable options.
// https://www.ffmpeg.org/ffmpeg-formats.html#hls-2
fmpeg('input.mp4', { timeout: 432000 }).addOptions([
    '-profile:v baseline', // baseline profile (level 3.0) for H264 video codec
    '-level 3.0', 
    '-s 640x360',          // 640px width, 360px height output video dimensions
    '-start_number 0',     // start the first .ts segment at index 0
    '-hls_time 10',        // 10 second segment duration
    '-hls_list_size 0',    // Maxmimum number of playlist entries (0 means all entries/infinite)
    '-f hls'               // HLS format
  ]).output('public/videos/output.m3u8').on('end', callback).run()

To create segments from an existing RTMP stream, use the following node-fluent-ffmpeg command. You can expect several seconds of latency, depending on hardware.

var ffmpeg = require('fluent-ffmpeg')

// host, port and path to the RTMP stream
var host = 'localhost'
var port = '1935'
var path = '/live/test'

function callback() { // do something when stream ends and encoding finshes }

fmpeg('rtmp://'+host+':'+port+path, { timeout: 432000 }).addOptions([
    '-c:v libx264',
    '-c:a aac',
    '-ac 1',
    '-strict -2',
    '-crf 18',
    '-profile:v baseline',
    '-maxrate 400k',
    '-bufsize 1835k',
    '-pix_fmt yuv420p',
    '-hls_time 10',
    '-hls_list_size 6',
    '-hls_wrap 10',
    '-start_number 1'
  ]).output('public/videos/output.m3u8').on('end', callback).run()

Using In-Memory Streams

By default, this module assumes files are kept in a directory on the local filesystem. If you want to stream files from another source (or don't want to relate URL paths to filesystem paths), you can specify a provider in the options like so:

var hls = new HLSServer(server, {
  provider: {
    exists: function (req, callback) { // check if a file exists (always called before the below methods)
      callback(null, true)                 // File exists and is ready to start streaming
      callback(new Error("Server Error!")) // 500 error
      callback(null, false)                // 404 error
    },
    getManifestStream: function (req, callback) { // return the correct .m3u8 file
      // "req" is the http request
      // "callback" must be called with error-first arguments
      callback(null, myNodeStream)
      // or
      callback(new Error("Server error!"), null)
    },
    getSegmentStream: function (req, callback) { // return the correct .ts file
      callback(null, myNodeStream)
    }
  }
})

See src/fsProvider.js for the default provider using the local filesystem.

CLI Tool

This package includes a CLI tool that can be installed globally with npm install -g hls-server.

To use, navigate to the directory where your .ts files are stored and run hlsserver in a command prompt. This will start a server on port 8000. (Use hlsserver --help to see additional options.)

The CLI tool will efficiently make use of multiple processors via the cluster module and can be used as an example of how to use the base module in the same way.

Notes

To publish from an RTMP client like OBS, use a RTMP server like rtmp-server-nodejs to echo the stream (direct streaming from that module is being worked on).

NOTE: Transcoding live streams is very CPU-intensive. Most consumer hardware won't be able to handle transcoding more than a few streams.

项目侧边栏1项目侧边栏2
推荐项目
Project Cover

豆包MarsCode

豆包 MarsCode 是一款革命性的编程助手,通过AI技术提供代码补全、单测生成、代码解释和智能问答等功能,支持100+编程语言,与主流编辑器无缝集成,显著提升开发效率和代码质量。

Project Cover

AI写歌

Suno AI是一个革命性的AI音乐创作平台,能在短短30秒内帮助用户创作出一首完整的歌曲。无论是寻找创作灵感还是需要快速制作音乐,Suno AI都是音乐爱好者和专业人士的理想选择。

Project Cover

白日梦AI

白日梦AI提供专注于AI视频生成的多样化功能,包括文生视频、动态画面和形象生成等,帮助用户快速上手,创造专业级内容。

Project Cover

有言AI

有言平台提供一站式AIGC视频创作解决方案,通过智能技术简化视频制作流程。无论是企业宣传还是个人分享,有言都能帮助用户快速、轻松地制作出专业级别的视频内容。

Project Cover

Kimi

Kimi AI助手提供多语言对话支持,能够阅读和理解用户上传的文件内容,解析网页信息,并结合搜索结果为用户提供详尽的答案。无论是日常咨询还是专业问题,Kimi都能以友好、专业的方式提供帮助。

Project Cover

讯飞绘镜

讯飞绘镜是一个支持从创意到完整视频创作的智能平台,用户可以快速生成视频素材并创作独特的音乐视频和故事。平台提供多样化的主题和精选作品,帮助用户探索创意灵感。

Project Cover

讯飞文书

讯飞文书依托讯飞星火大模型,为文书写作者提供从素材筹备到稿件撰写及审稿的全程支持。通过录音智记和以稿写稿等功能,满足事务性工作的高频需求,帮助撰稿人节省精力,提高效率,优化工作与生活。

Project Cover

阿里绘蛙

绘蛙是阿里巴巴集团推出的革命性AI电商营销平台。利用尖端人工智能技术,为商家提供一键生成商品图和营销文案的服务,显著提升内容创作效率和营销效果。适用于淘宝、天猫等电商平台,让商品第一时间被种草。

Project Cover

AIWritePaper论文写作

AIWritePaper论文写作是一站式AI论文写作辅助工具,简化了选题、文献检索至论文撰写的整个过程。通过简单设定,平台可快速生成高质量论文大纲和全文,配合图表、参考文献等一应俱全,同时提供开题报告和答辩PPT等增值服务,保障数据安全,有效提升写作效率和论文质量。

投诉举报邮箱: service@vectorlightyear.com
@2024 懂AI·鲁ICP备2024100362号-6·鲁公网安备37021002001498号