Recherche avancée

Sur d’autres sites (8908)

• Run python ffmpeg audio convertion code file from subprocess.call() within a flask server

  11 novembre 2019, par Kasun

  I have build a small flask server to handle the request. I have 3 parameters in the api function that i want to get. those are type, user_id, audio_file, One is a file. Since it’s used for the audio file conversion. I have to get a file.. I have tested with this in Postman audio file get saved but the subprocess.call(command) in the api function doesn’t work..

  this is the flask server code

  @app.route('/voice', methods=['GET','POST'])
  
  def process_audio():
  
   try:
  
      if request.method == 'POST':
  
          input_type = request.form['type']
  
          user_id = request.form['user_id']
  
          static_file = request.files['audio_file']
  
          audio_name = secure_filename(static_file.filename)
  
          path = 't/'
  
          status = static_file.save(path + audio_name)
  
          full_file = path + audio_name
  
          if input_type == 1:
  
              cmd = "python t/convert_english.py --audio " + full_file
  
              res = subprocess.call([cmd],shell=True)
  
              f = open('t/ress.txt', 'w')
  
              f.write(str(res))
  
              f.close()
  
              return "true"
  
          else:
  
              cmd = "python t/convert_sinhala.py --audio " + full_file
  
              os.system(cmd)
  
              return "true"
  
      else:
  
          return "false"
  
  
  
  except Exception as e:
  
      print(e)
  
      logger.error(e)
  
      return e

  The audio file get saved in the directory as expected..

  this is the convert_english.py

  import subprocess
  
  import argparse
  
  import os
  
  import logging
  
  import speech_recognition as sr
  
  from tqdm import tqdm
  
  from multiprocessing.dummy import Pool
  
  
  
  #subprocess.call('pip install pydub',shell=True)
  
  
  
  from os import path
  
  from pydub import AudioSegment
  
  
  
  logging.basicConfig(filename='/var/www/img-p/t/ee.log', level=logging.DEBUG, 
  
                      format='%(asctime)s %(levelname)s %(name)s %(message)s')
  
  logger=logging.getLogger(__name__)
  
  
  
  ap = argparse.ArgumentParser()
  
  ap.add_argument("-a", "--audio", required=True,
  
      help="path to input audio file")
  
  args = vars(ap.parse_args())
  
  
  
  src = args["audio"]
  
  dst = "audio.wav"
  
  
  
  sound = AudioSegment.from_mp3(src)
  
  sound.export(dst, format="wav")
  
  
  
  #subprocess.call('pip install ffmpeg-python',shell=True)
  
  
  
  subprocess.call('mkdir parts',shell=True)
  
  
  
  subprocess.call('ffmpeg -i audio.wav -f segment -segment_time 30 -c copy parts/out%09d.wav',shell=True)
  
  
  
  #subprocess.call('pip install SpeechRecognition',shell=True)
  
  
  
  
  
  pool = Pool(8) # Number of concurrent threads
  
  
  
  with open("api-key.json") as f:
  
      GOOGLE_CLOUD_SPEECH_CREDENTIALS = f.read()
  
  
  
  r = sr.Recognizer()
  
  files = sorted(os.listdir('parts/'))
  
  
  
  def transcribe(data):
  
      idx, file = data
  
      name = "parts/" + file
  
      print(name + " started")
  
      # Load audio file
  
      with sr.AudioFile(name) as source:
  
          audio = r.record(source)
  
      # Transcribe audio file
  
      text = r.recognize_google_cloud(audio, credentials_json=GOOGLE_CLOUD_SPEECH_CREDENTIALS)
  
      print(name + " done")
  
      return {
  
          "idx": idx,
  
          "text": text
  
      }
  
  
  
  all_text = pool.map(transcribe, enumerate(files))
  
  pool.close()
  
  pool.join()
  
  
  
  transcript = ""
  
  for t in sorted(all_text, key=lambda x: x['idx']):
  
      total_seconds = t['idx'] * 30
  
      # Cool shortcut from:
  
      # https://stackoverflow.com/questions/775049/python-time-seconds-to-hms
  
      # to get hours, minutes and seconds
  
      m, s = divmod(total_seconds, 60)
  
      h, m = divmod(m, 60)
  
  
  
      # Format time as h:m:s - 30 seconds of text
  
      transcript = transcript + "{:0>2d}:{:0>2d}:{:0>2d} {}\n".format(h, m, s, t['text'])
  
  
  
  print(transcript)
  
  
  
  with open("transcript.txt", "w") as f:
  
      f.write(transcript)
  
  
  
  f = open("transcript.txt")
  
  lines = f.readlines()
  
  f.close()
  
  f = open("transcript.txt", "w", encoding="utf-8")
  
  for line in lines:
  
      f.write(line[8:])
  
  f.close()

  The thing is above code works when i manually run the command -> python t/convert_english.py —audio t/tttttttttt.mp3 like this in the terminal..

  But when i try to run from the flask server itself it doesn’t works.. And I’m not getting an error either.

• How to fetch video frame and its timestamp from ffmpeg to python code

  14 février 2017, par vijiboy

  Searching for an alternative as OpenCV would not provide timestamps which were required in my computer vision algorithm, I found this excellent article https://zulko.github.io/blog/2013/09/27/read-and-write-video-frames-in-python-using-ffmpeg/
  Working up the code on windows I still could’nt get the frame timestamps.
  I recollected seeing on ffmpeg forum somewhere that the video filters like showinfo are bypassed when redirected.

  Here’s what I tried :

  import subprocess as sp
  
  import numpy
  
  import cv2
  
  
  
  command = [ 'ffmpeg', 
  
              '-i', 'e:\sample.wmv',
  
              '-pix_fmt', 'rgb24',
  
              '-vcodec', 'rawvideo',
  
              '-vf', 'showinfo', # video filter - showinfo will provide frame timestamps
  
              '-an','-sn', #-an, -sn disables audio and sub-title processing respectively
  
              '-f', 'image2pipe', '-'] # we need to output to a pipe
  
  
  
  pipe = sp.Popen(command, stdout = sp.PIPE, stderr = sp.STDOUT) # TODO someone on ffmpeg forum said video filters (e.g. showinfo) are bypassed when stdout is redirected to pipes??? 
  
  
  
  for i in range(10):
  
      raw_image = pipe.stdout.read(1280*720*3)
  
      img_info = pipe.stdout.read(244) # 244 characters is the current output of showinfo video filter
  
      print "showinfo output", img_info
  
      image1 =  numpy.fromstring(raw_image, dtype='uint8')
  
      image2 = image1.reshape((720,1280,3))  
  
  
  
      # write video frame to file just to verify
  
      videoFrameName = 'Video_Frame{0}.png'.format(i)
  
      cv2.imwrite(videoFrameName,image2)
  
  
  
      # throw away the data in the pipe's buffer.
  
      pipe.stdout.flush()

  So how to still get the frame timestamps from ffmpeg into python code so that it can be used in my computer vision algorithm ...

• Anomalie #3620 : Token de prévisualisation inconsistant selon que l’article est en cours de rédact...

  12 février 2017

  b b a écrit :

  Donc, je parle bien d’une fonctionnalité native de SPIP dont le comportement a changé entre SPIP 2.1 et SPIP 3.1

  Non, il n’y a pas non plus de trace de var_relecture dans le core en 2.1, la fonctionnalité du var_relecture est dans le plugin bonux et diffère du var_mode preview du core.

  Si j’ai donné les fichiers concernés, c’est bien pour montrer que var_relecture ne faisait jamais partie du Core ;-)

  Contrairement au token pour la preview qui lui, est disponible depuis 3 ans en 2.1 (cf r21084)