Recherche avancée

Sur d’autres sites (8715)

• How do you determine the end of the file in a stream containing multiple streams ? (nodejs)

  26 décembre 2019, par Danielkent

  I would like to split an audio file into multiple segments using ffmpeg in an AWS Lambda (NodeJS) function.

  Due to the limitations of (and to optimise for) the lambda environment I would like to stream the audio into ffmpeg, perform the split on the audio file in the stream and then stream the now multiple smaller files out to s3.

  After doing some research I have found the AWS S3 SDK doesn’t support multiple file uploads in one stream. I could resolve this by finding the end of each new segment (file in the output stream) and creating a separate upload to s3.

  Is there a way to determine the end of a file in a stream (containing multiple files) ?

  (without saving it to the file system or loading it to memory).

  I have searched around and I can’t seem to find an answer.

• Google Speech API - Is there a way to determine if the audio has human voice or not ?

  20 décembre 2019, par stupid_sma

  I am making an audio filtering application at work that reads over hundreds of audio files and filters them. So, if the audio has human voice in it, it will accept it and if it does not- it will delete the audio file.

  I am using ffmpeg to get the details of the audio and add other filters like size and duration and silence (though it is not very accurate in detecting silence for all audio files.)

  My company asked me to try the Google Cloud Speech API to detect if the audio has any human voice in it.

  So with this code, some audio files return a Transcript of spoken words in the audio file, but what I need is to determine if a human is speaking or not.

  I have considered using hark.js for it but there does not seem to be enough documentation and I am short on time !

  Ps. I am an intern and I’m just starting out with programming. I apologize if my question does not make sense or sounds dumb.

     # Includes the autoloader for libraries installed with composer
  
     require __DIR__ . '/vendor/autoload.php';
  
  
  
     # Imports the Google Cloud client library
  
     use Google\Cloud\Speech\V1\SpeechClient;
  
     use Google\Cloud\Speech\V1\RecognitionAudio;
  
     use Google\Cloud\Speech\V1\RecognitionConfig;
  
     use Google\Cloud\Speech\V1\RecognitionConfig\AudioEncoding;
  
  
  
     putenv('GOOGLE_APPLICATION_CREDENTIALS=../../credentials.json');
  
  
  
  
  
  
  
     echo getcwd() . "<br />";
  
     chdir('test-sounds');
  
     echo getcwd() . "<br />";
  
     echo shell_exec('ls -lr');
  
  
  
     $fileList = glob('*');
  
     foreach($fileList as $filename){
  
     //echo $filename, '<br />';
  
  
  
     # The name of the audio file to transcribe
  
     $audioFile = __DIR__ . '/' . $filename;
  
  
  
     # get contents of a file into a string
  
     $content = file_get_contents($audioFile);
  
  
  
     # set string as audio content
  
     $audio = (new RecognitionAudio())
  
         ->setContent($content);
  
  
  
     # The audio file's encoding, sample rate and language
  
     $config = new RecognitionConfig([
  
         'encoding' => AudioEncoding::LINEAR16,
  
         'language_code' => 'ja-JP'
  
     ]);
  
  
  
     # Instantiates a client
  
     $client = new SpeechClient();
  
  
  
     # Detects speech in the audio file
  
     $response = $client->recognize($config, $audio);
  
  
  
     # Print most likely transcription
  
     foreach ($response->getResults() as $result) {
  
         $alternatives = $result->getAlternatives();
  
         $mostLikely = $alternatives[0];
  
         $transcript = $mostLikely->getTranscript();
  
         printf('<br />Transcript: %s' . PHP_EOL, $transcript . '<br />');
  
  
  
     }
  
  
  
     $client->close();
  
  
  
     }
  
  
  
     ?> ```

• Is it possible to determine if a subtitle track is imaged based or text based with ffprobe

  21 février 2021, par Shex

  I'm writing a script that burns subtitles into video files to prepare them for a personal stream I'm hosting. I'm having a hard time finding which type of subtitle is used in the file. I use ffprobe to get the files' information, and I can get stuff like the codec type, but I was wondering if there is a way to determine if a subtitle track is image based or text based. I can only think of getting a list of all possible codecs and match the codec type with this list but it would be very useful to have an info somewhere that can tell me "OK this is an image-based subtitle track", as when I burn I cannot use the same filters with ffmpeg to burn image vs. text subtitles.