Recherche avancée

Sur d’autres sites (4778)

• How to use ffmpeg to transcode many live streamed videos ? [closed]

  21 septembre 2020, par user14258924

  PREMISE

  


  As a pet project, I am writing a live video streaming service, in Go, that can consume video streams from OBS via SRT(TS -> h264/aac) and RTMP(FLV -> h264/aac) protocols and am planning to support streaming video from web browser as well, captured from a web camera via JS. This ingress server will receive many video streams in various containers and codecs and I need to normalize them into single container and codec and then create multiple versions for various bitrates(ie. 240p, 360p, 480p, 720p, 1080p...) to pass along where needed in the application. Each stream is split into 2 second GOP segments, separate for audio and video track, that will produce fragmented MP4 as the end result - which can be consumed by web browser.

  


  The issue is that I am using Go which has no libraries for transcoding video so I need to use either ffmpeg or vlc, which is a C code. I have decided to avoid the CGo route and use ffmpeg/vlc as standalone binaries.

  


  QUESTION

  


  My question is how to use either of these project in the most efficient way - avoiding the use of files in favour of unix sockets/streams and also the performance aspect - handling hundreds of video segments in any one time and in sufficient time to avoid creating too much of a lag beteen producer and consumer.

  


  So let's say I will pick the most used one - ffmpeg, how should I actually use it to achieve what I have described ? How would you set it up and which flags/config to use with it ?

  


  Can the performance be even achieved or is it just too much and I will need some sort of ffmpeg cluser to even come close to some useful performance/low delay ?

  


• Merging input Streams with nodejs/ffmpeg

  14 septembre 2020, par jAndy

  I'm creating a very basic and rudimentary Video-Web-Chat. On the client side, I'm going to use a simple getUserMedia API call to capture the webcam data and send video-data as data-blob to my server.

  


  From There, I'm planning to either use the fluent-ffmpeg library or just spawn ffmpeg myself and pipe that raw data to ffmpeg, which in turn, does some magic and pushes that out as HLS stream to an Amazon AWS Service (for instance), which then gets actually displayed on a Web Browser for all participating people in the video chat.

  


  So far, I think all of this should be fairly easy to implement, but I keep my head spinning around the question, how I can create a "combined" or "merged" frame and stream, so the output HLS data from my server to the distributing cloud service has only to be one combined data stream to receive.

  


  If there are 3 people in that video chat, my server receives 3 data streams from those clients and combines these data streams (from the individual web-cam data sources) into one output stream.

  


  How could that be accomplished ?
Can I "create" a new frame with ffmpeg, so to speak ? I would be very thankful if anybody could give me a heads up here, maybe I'm thinking in a complete wrong direction.

  


  Another question which arises to me is, if I really can just "dump" any data, which I'm receiving from a binary blob created from getUserMedia or MultiStreamRecorder to ffmpeg or if I have to specify somewhere and somehow the exact codecs being used etc.?

  


• creating illusion of live streaming (internet radio) using ffmpeg

  29 décembre 2014, par user259060

  I am working on a project that involves live streaming but without seeking (just like internet radio). I am using ffmpeg and ffserver.

  • I could just send the song to ffserver feed using ffmpeg but the problem is that the whole song / file is getting dumped. I don’t want that to happen.

  • First I segmented the song using ffmpeg -threads 1 -i INPUT.mp3 -ar 24000 -acodec libmp3lame -ac 1 -aq 1 -ab 64k -map 0:0 -f segment -segment_time 2 -segment_list /PATH/TO/LIST/outputlist.m3u8 -segment_format mpegts /OUTPUT/PATH/output%05d.mp3 (this is just an example) .

  • As you can see that the segment time is 2 seconds. What I’m actually planning to achieve is that I want to send first segmented file say output00001.mp3 to ffserver feed and wait for 1 second then send the the second segmented files say output00002.mp3 to ffserver feed and so on till the end of the song. This creates an illusion of radio live streaming. I was able to implement this without a problem using python.

  PROBLEM :

  The problem I faced while listening to song (htttp ://foo:port/test1.mp3) is that after 2nd second I hear a few millisecond pause and then the song continues to play which is very irritating. This happens after every segmented song completes playing. Is there any solution to eradicate the pause ? Is there a technique to make song play live (that means if my server shuts down, the song should stop immediately) ?