Recherche avancée

Sur d’autres sites (11102)

• FFMpeg-Normalize Doesn't Exist After Install

  26 octobre 2019, par 123hierarchy

  I’m cloning this repo. I installed ffmpeg-normalize and pip install ffmpeg-normalize returns

  Requirement Already Satisfied...

  And yet I’m getting an error that ffmpeg-normalize doesn’t exist when executing the preprocess.py step :

  

  The offending code :

  def preprocess(audio_filename, output_filename):
  
      ext_ind = audio_filename.rfind('.wav')
  
      audio_filename_formatted = audio_filename[:ext_ind] + '-formatted.wav'
  
      try:
  
          os.remove(audio_filename_formatted)
  
      except OSError:
  
          pass
  
      try:
  
          os.remove(output_filename)
  
      except OSError:
  
          pass
  
      error = os.system(
  
          'ffmpeg -i {} -acodec pcm_s16le -ac 1 -ar 16000 {}'.format(
  
              audio_filename, audio_filename_formatted))
  
      if error:
  
          print error
  
          raise StandardError('ffmpeg or audio file doesn\'t exist')
  
      error = os.system(
  
          'ffmpeg-normalize -f {}'.format(audio_filename_formatted))
  
      if error:
  
          raise StandardError('ffmpeg-normalize doesn\'t exist')
  
  
  
      data = wavfile.read(audio_filename_formatted)
  
      mfcc_inst = MFCC()
  
      features = mfcc_inst.sig2s2mfc_energy(data[1])
  
  
  
      np.save(output_filename, features)
  
  
  
  
  
  if __name__ == '__main__':
  
      if len(sys.argv) < 3:
  
          sys.exit(
  
              "Have to pass audio_filename and output_filename as parameters.")
  
      print sys.argv[1],sys.argv[2]
  
      preprocess(sys.argv[1], sys.argv[2])

  Is there another step after installing it that’s needed ? Is this an issue with running it on a Mac ? I’m completely lost here. Thank you in advance.

• Can't upload huge video to google storage. I using "@ffmpeg-installer/ffmpeg" and @google-cloud/storage

  20 juillet 2022, par Dmytro Petskovych

  I upload file to google storage using "@ffmpeg-installer/ffmpeg" and @google-cloud/storage in my node.js App.
Step 1. file uploading to fs is in child processes - one process for each type of resolution (totaly six).
step 2. encription (converting to stream)
step 3. upload to google storage

  


  I use "Upload a directory to a bucket" in order to send the video from the client to the Google Cloud Storage bucket.

  


  This way is working fine only with small file.

  


  when I upload video, actually I upload six videos, one for each type resolution

  


  for example when I upload video with duration one hour it split on chunk and totally I get more three thousands files.

  


  So actually i upload folder with large amount of files, but not all of this files are uploaded to cloud.

  


  maybe someone had the similar problem and helps fix it.

  


  


  


  const uploadFolder = async (bucketName, directoryPath, socketInstance) => {
    try {
      let dirCtr = 1;
      let itemCtr = 0;
      const fileList = [];

      const onComplete = async () => {
        const folderName = nanoid(46);

        await Promise.all(
          fileList.map(filePath => {
            const fileName = path.relative(directoryPath, filePath);
            const destination = `${ folderName }/${ fileName }`;

            return storage
              .bucket(bucketName)
              .upload(filePath, { destination })
              .then(
                uploadResp => ({ fileName: destination, status: uploadResp[0] }),
                err => ({ fileName: destination, response: err })
              );
          })
        );

        if (socketInstance) socketInstance.emit('uploadProgress', {
          message: `Added files to Google bucket`,
          last: false,
          part: false
        });

        return folderName;
      };

      const getFiles = async directory => {
        const items = await fs.readdir(directory);
        dirCtr--;
        itemCtr += items.length;
        for(const item of items) {
          const fullPath = path.join(directory, item);
          const stat = await fs.stat(fullPath);
          itemCtr--;
          if (stat.isFile()) {
            fileList.push(fullPath);
          } else if (stat.isDirectory()) {
            dirCtr++;
            await getFiles(fullPath);
          }
        }
      }

      await getFiles(directoryPath);

      return onComplete();
    } catch (e) {
      log.error(e.message);
      throw new Error('Can\'t store folder.');
    }
  };

  


  


  



• Merge audio (m4s) segments into one

  20 avril 2022, par akinuri

  I recently started learning Laravel, and currently watching an online course. Online courses are fine, but I like to have local copies, so I'm trying to download/merge segmented audio from Laracasts : Laravel 8 From Scratch series.

  


  I've written some scripts (in Python) that does the following :

  


  

  1. Download the master.json

  


  2. Read master.json and download audio segments

  


  3. Merge the segments into a single file (the file is not playable yet)

  


  4. Process the audio file via ffmpeg (now it's playable, but has issues)

  


  


  I think there's a problem with the step 3 and/or 4.

  


  In step/script 3, I create a new file, and add the contents of the segments to the file in binary.

  


  Then (step/script 4), run a ffmpeg command in python : ffmpeg -i merged-file.mp4 -c copy processed-file.mp4

  


  However, the final file doesn't work/play as expected. There's a delay in the beginning, and some parts seem to be cut off/skipped.

  


  There are three possibilities :

  


  

  • Segment files are problematic (not likely ?)

  


  • I'm doing the merging wrong

  


  • I'm doing the ffmpeg processing wrong

  


  


  Can someone guide me here ?

  


  ---

  


  The issues/colored parts in the ffmpeg output are :

  


  ...
[mov,mp4,m4a,3gp,3g2,mj2 @ 000001cfbc0de780] could not find corresponding track id 2
[mov,mp4,m4a,3gp,3g2,mj2 @ 000001cfbc0de780] could not find corresponding trex (id 2)
...
[aac @ 000001cfbc0f0380] Number of bands (31) exceeds limit (6).
...
[mp4 @ 000001cfbc20ecc0] track 0: codec frame size is not set
...
[mp4 @ 000001cfbc20ecc0] Non-monotonous DTS in output stream 0:0; previous: 318318, current: 286286; changing to 318319. This may result in incorrect timestamps in the output file.
...


  


  Everything required for a test case is located in GitHub (akinuri/dump/m4s-segments/). Screenshot of the contents :

  


  

  


  ---

  


  Note : there are two types/formats of audio in the master.json : mp42 and dash. dash works as expected, and seem to be used in limited videos/courses. On the other hand, mp42 appears more. So I need a way to make mp42 work.