Recherche avancée

Sur d’autres sites (9934)

• FFmpeg wrong codecstring for DASH manifest [closed]

  21 novembre 2024, par Suxsem

  I'm trying to create a DASH stream with ffmpeg with the following command :

  


  ffmpeg -i rtsp://admin:***@camera-retro.lan/ch0_0 ^
       -map 0 ^
       -codec:v copy ^
       -codec:a copy ^
       -f dash ^
       -use_template 1 ^
       -use_timeline 1 ^
       -window_size 50 ^
       -extra_window_size 5 ^
       -seg_duration 5 ^
       tmp/output.mpd


  


  The problem is the generated manifest doesn't contain a valid codecstring for the video part (note the codecs="hev1" part) and thus is not playable by the browser :

  


  &lt;?xml version="1.0" encoding="utf-8"?>&#xA;<mpd xmlns="urn:mpeg:dash:schema:mpd:2011" profiles="urn:mpeg:dash:profile:isoff-live:2011" type="dynamic" minimumupdateperiod="PT6S" suggestedpresentationdelay="PT6S" availabilitystarttime="2024-11-21T12:05:52.604Z" publishtime="2024-11-21T12:05:57.559Z" timeshiftbufferdepth="PT5M1.9S" maxsegmentduration="PT5.0S" minbuffertime="PT12.0S">&#xA;    <programinformation>&#xA;        &#xA;    </programinformation>&#xA;    <servicedescription>&#xA;    </servicedescription>&#xA;    <period start="PT0.0S">&#xA;        <adaptationset contenttype="video" startwithsap="1" segmentalignment="true" bitstreamswitching="true" maxwidth="2304" maxheight="1296" par="16:9">&#xA;            <representation mimetype="video/mp4" codecs="hev1" bandwidth="822093" width="2304" height="1296" scantype="unknown" sar="1:1">&#xA;                <segmenttemplate timescale="90000" initialization="init-stream$RepresentationID$.m4s" media="chunk-stream$RepresentationID$-$Number%05d$.m4s" startnumber="1">&#xA;                    <segmenttimeline>&#xA;                        <s t="0" d="542990"></s>&#xA;                    </segmenttimeline>&#xA;                </segmenttemplate>&#xA;            </representation>&#xA;        </adaptationset>&#xA;        <adaptationset contenttype="audio" startwithsap="1" segmentalignment="true" bitstreamswitching="true">&#xA;            <representation mimetype="audio/mp4" codecs="mp4a.40.2" bandwidth="34143" audiosamplingrate="16000">&#xA;                <audiochannelconfiguration schemeiduri="urn:mpeg:dash:23003:3:audio_channel_configuration:2011" value="1"></audiochannelconfiguration>&#xA;                <segmenttemplate timescale="16000" initialization="init-stream$RepresentationID$.m4s" media="chunk-stream$RepresentationID$-$Number%05d$.m4s" startnumber="1">&#xA;                    <segmenttimeline>&#xA;                        <s t="0" d="81001"></s>&#xA;                    </segmenttimeline>&#xA;                </segmenttemplate>&#xA;            </representation>&#xA;        </adaptationset>&#xA;    </period>&#xA;</mpd>&#xA;

  


  This is the output of ffprobe :

  


  SDP:
v=0
o=- 1732188474342789 1 IN IP4 192.168.12.162
s=Session streamed by "rRTSPServer"
i=ch0_0.h264
t=0 0
a=tool:LIVE555 Streaming Media v2023.01.19
a=type:broadcast
a=control:*
a=range:npt=now-
a=x-qt-text-nam:Session streamed by "rRTSPServer"
a=x-qt-text-inf:ch0_0.h264
m=video 0 RTP/AVP 96
c=IN IP4 0.0.0.0
b=AS:700
a=rtpmap:96 H265/90000
a=fmtp:96 profile-space=0;profile-id=1;tier-flag=0;level-id=186;interop-constraints=000000000000;sprop-vps=QAEMAf//AWAAAAMAAAMAAAMAAAMAuqwJ;sprop-sps=QgEBAWAAAAMAAAMAAAMAAAMAuqABICAFEf5a7kSIi/Lc1AQEBAI=;sprop-pps=RAHA8oSJAzJA
a=control:track1
m=audio 0 RTP/AVP 97
c=IN IP4 0.0.0.0
b=AS:32
a=rtpmap:97 MPEG4-GENERIC/16000
a=fmtp:97 streamtype=5;profile-level-id=1;mode=AAC-hbr;sizelength=13;indexlength=3;indexdeltalength=3;config=1408
a=control:track2


  


  it seems to me that all codec informations are present :

  


  a=fmtp:96 profile-space=0;profile-id=1;tier-flag=0;level-id=186;interop-constraints=000000000000;


  


  why ffmpeg is putting only codecs="hev1" instead of the full codecstring (containing the profile, the level and the constraints) ?

  


  Thank you

  


• RTP stream from ffmpeg not reaching mediasoup producer : what could be wrong ? [closed]

  5 novembre 2024, par Francesco Vignola

  I developed a producer for mediasoup that is supposed to receive the RTP stream generated by ffmpeg (via fluent-ffmpeg) and send it to the consumer.

  


  ffmpeg correctly generates the stream, and I conducted a test with tcpdump, where I can see the traffic generated by ffmpeg being sent locally to the port of the PlainTransport with which the producer is created.

  


  However, neither the producer nor the consumer seem to be able to intercept it, as the listeners I attached to the specific events are never triggered.

  


  I also started an ffmpeg client to receive the stream, and the test was successful (it receives the stream). However, I can’t understand why it doesn't work with mediasoup.

  


  Can someone help me ? Did I forget any configuration parameters ?

  


  The code is as follows :

  


  const router = await worker.createRouter({
   mediaCodecs: [{
      kind: 'video',
      mimeType: 'video/VP8',
      clockRate: 90000,
      payloadType: 96
   }]
});

const transport = await router.createPlainTransport({
   listenIp: { ip: '127.0.0.1' },
   rtcpMux: false,
   comedia: true
});

const producer = await transport.produce({
   kind: 'video',
   rtpParameters: router.rtpParameters
});

const { localIp: ip, localPort: port } = transport.tuple;
const { localPort: rtcpPort } = transport.rtcpTuple;

await transport.connect({
   ip: ip,
   port: port,
   rtcpPort: rtcpPort
});

ffmpeg(videoFilePath)
   .inputOptions('-re')
   .videoCodec('libvpx')
   .outputOptions([
       '-f rtp',
       '-payload_type', '96',
       '-ssrc', ssrc,
       '-an',
       '-sdp_file', 'out.sdp'
   ])
   .output(`rtp://${ip}:${port}?rtcpport=${rtcpPort}`)
   .run();


  


  The video to be streamed has an .mp4 extension and is saved locally on the server.

  


  Thank you to anyone who is willing to help me.

  


• Audio offset get wrong after some time when streaming audios

  7 septembre 2024, par Antoine Grenard

  I use microsoft-cognitiveservices-speech-sdk (1.38.0) in order to do real time speech to text.
It seems like the offset is right when I send a full audio but it is wrong when I send it cut in a lot of audio chunks.

  


  The more there is audio chunks the more inaccurate the offset is :

  


  

  • No chunks : 1 726 300 000

  


  • 369 chunks of 0.5 seconds : 1 729 600 000

  


  • 923 chunks of 0.2 seconds : 1 744 600 000

  


  • 1443 chunks of 0.1 seconds : 1 757 900 000

  


  


  To reproduce here is some piece of code :

  


      const speechConfig = SpeechConfig.fromSubscription(<key>,  {console.log(event)}&#xA;    speechRecognizer.canceled = async (recognizer, event) => {console.log(event)}&#xA;    speechRecognizer.startContinuousRecognitionAsync();&#xA;&#xA;    for (let i = 1; i &lt;= 1443; i&#x2B;&#x2B;) {&#xA;      const formattedNumber = i.toString().padStart(4, &#x27;0&#x27;);&#xA;      const buffer = fs.readFileSync(`/var/tmp/chunks/output_${formattedNumber}.wav`);&#xA;      pushStream.write(buffer);&#xA;    }&#xA;</key>

  


  To create the audio chunks :

  


  ffmpeg -i  -f segment -segment_time 0.1 -c copy output_%04d.wav


  


  Here is the audio link : https://drive.google.com/file/d/1H_RJuqMiBaVkpo9XHrgp1bpuFdgQl64O/view?usp=sharing

  


  Thanks for your help