Recherche avancée

Sur d’autres sites (6562)

• FFmpeg compiling error(libavfilter/vf_scale_cuda.cu)

  11 août 2021, par Amir Mousavi

  installed packages :

  


  


  libx264 - cuda11.2 - cuda-toolkit - nvidia-driver-465

  


  


  command :

  


  ./configure —enable-nonfree —enable-cuvid —enable-nvenc —enable-cuda-nvcc —enable-libx264 —enable-gpl —extra-cflags=-I/usr/local/cuda/include —extra-ldflags=-L/usr/local/cuda/lib64

  


  here is latest lines errors :

  


  


  libavfilter/vf_scale_cuda.cu(1123) : error : a template argument may not reference a non-external entity

  


  libavfilter/vf_scale_cuda.cu(1123) : error : a template argument may not reference a non-external entity

  


  libavfilter/vf_scale_cuda.cu(1124) : error : a template argument may not reference a non-external entity

  


  libavfilter/vf_scale_cuda.cu(1124) : error : a template argument may not reference a non-external entity

  


  libavfilter/vf_scale_cuda.cu(1124) : error : a template argument may not reference a non-external entity

  


  libavfilter/vf_scale_cuda.cu(1124) : error : a template argument may not reference a non-external entity

  


  Error limit reached.
100 errors detected in the compilation of >"/tmp/tmpxft_000032a1_00000000-7_vf_scale_cuda.cpp1.ii".
Compilation terminated.
ffbuild/common.mak:108 : recipe for target >'libavfilter/vf_scale_cuda.ptx' failed
make : *** [libavfilter/vf_scale_cuda.ptx] Error 4
make : *** Waiting for unfinished jobs....

  


  


  I don't know where is the problem.

  


  lines that error referenced :

  


  EAREST_KERNELS(yuv420p)
NEAREST_KERNELS(nv12)
NEAREST_KERNELS(yuv444p)
NEAREST_KERNELS(p010le)
NEAREST_KERNELS(p016le)
NEAREST_KERNELS(yuv444p16le)


  


• OpenAI Whisper : FileNotFoundError : [WinError 2] The system cannot find the file specified

  16 décembre 2023, par Alex

  I'm not the first to get this error, however, all previous answers relate to ffmpeg not installed. As you can see, I installed ffmpeg so not sure why i still get this error. Moreover, this code works perfectly fine in Google Colab but not in Spyder.

  


  pip install tiktoken

pip install -U openai-whisper

pip install ffmpeg

pip install setuptools-rust

import whisper

model = whisper.load_model("medium")

result = model.transcribe("F:/iim/dahl/01.Projects/20231215_Schlueter/02.Code/01.Data/The Investors Everybody Ignored ft Sallie Krawcheck  Whats Your Problem  Jacob Goldstein.mp3")
print(result["text"])



 Cell In[13], line 1
    result = model.transcribe("F:/iim/dahl/01.Projects/20231215_Schlueter/02.Code/01.Data/The Investors Everybody Ignored ft Sallie Krawcheck  Whats Your Problem  Jacob Goldstein.mp3/")

  File F:\iim\dahl\00.Anaconda\Lib\site-packages\whisper\transcribe.py:122 in transcribe
    mel = log_mel_spectrogram(audio, model.dims.n_mels, padding=N_SAMPLES)

  File F:\iim\dahl\00.Anaconda\Lib\site-packages\whisper\audio.py:140 in log_mel_spectrogram
    audio = load_audio(audio)

  File F:\iim\dahl\00.Anaconda\Lib\site-packages\whisper\audio.py:58 in load_audio
    out = run(cmd, capture_output=True, check=True).stdout

  File F:\iim\dahl\00.Anaconda\Lib\subprocess.py:548 in run
    with Popen(*popenargs, **kwargs) as process:

  File F:\iim\dahl\00.Anaconda\Lib\site-packages\spyder_kernels\customize\spydercustomize.py:109 in __init__
    super(SubprocessPopen, self).__init__(*args, **kwargs)

  File F:\iim\dahl\00.Anaconda\Lib\subprocess.py:1026 in __init__
    self._execute_child(args, executable, preexec_fn, close_fds,

  File F:\iim\dahl\00.Anaconda\Lib\subprocess.py:1538 in _execute_child
    hp, ht, pid, tid = _winapi.CreateProcess(executable, args,

FileNotFoundError: [WinError 2] Das System kann die angegebene Datei nicht finden



  


• Low Latency DASH Nginx RTMP

  10 mai 2017, par Kenan Christian

  I use arut nginx-rtmp-module (https://github.com/arut/nginx-rtmp-module) on the media server, then I tried to stream using FFmpeg to the dash application, then I test the stream by playing it using VLC.

  And it waits around 30secs to start playing, and it plays from the beginning, not the current timestamp.

  This is my current config on the RTMP block

  rtmp {
  
      server {
  
          listen 1935;
  
  
  
          application live {
  
              live on;
  
  
  
             exec ffmpeg -re -i rtmp://localhost:1935/live/$name
  
                -c:a libfdk_aac -b:a 32k  -c:v libx264 -b:v 128K -f flv rtmp://localhost:1935/hls/$name_low
  
                -c:a libfdk_aac -b:a 64k  -c:v libx264 -b:v 256k -f flv rtmp://localhost:1935/hls/$name_mid
  
                -c:a libfdk_aac -b:a 128k -c:v libx264 -b:v 512K -f flv rtmp://localhost:1935/hls/$name_hi
  
                -c:a libfdk_aac -b:a 128k -c:v libx264 -b:v 512K -f flv rtmp://localhost:1935/dash/$name_dash;
  
          }
  
  
  
          application hls {
  
               live on;
  
  
  
               hls on;
  
               hls_path /tmp/hls;
  
               hls_nested on;
  
  
  
               hls_variant _low BANDWIDTH=160000;
  
               hls_variant _mid BANDWIDTH=320000;
  
               hls_variant _hi  BANDWIDTH=640000;
  
          }
  
  
  
          application dash {
  
              live on;
  
  
  
              dash on;
  
              dash_path /tmp/dash;
  
              dash_nested on;
  
          }
  
      }
  
  }

  This is the command I use for streaming

  ffmpeg -re -i 2014\ SPRING.mp4 -c copy -f flv 
  
  rtmp://52.221.221.163:1935/dash/spring

  How can I reduce the delay, and make it play from the same timestamp as the streamer ?

  Can I achieve under 5s latency ?

  UPDATE

  Tried to change the playlist length and fragment length, using this directive

  dash_playlist_length 10s;
  
  dash_fragment 2s;

  But still got some latency problem, sometimes it’s smaller than before, sometimes it’s the same