1

app.py

@@ -7,6 +7,8 @@ from pydub import AudioSegment
 # Import everything needed to edit video clips
 from moviepy.editor import *
 import numpy as np
+import mutagen
+from mutagen.wave import WAVE
 
 #path = "/content/gdrive/My Drive/AI/sample_imgs/"
 #dirs = os.listdir( path )
@@ -55,17 +57,19 @@ def merge_audio_video(fps, resize_img_list, speech):
     
     #convert flac to mp3 audio format 
     wav_audio = AudioSegment.from_file(speech, "flac")  #("/content/gdrive/My Drive/AI/audio1.flac", "flac")
+    print('flac audio read', type(wav_audio))
     wav_audio.export("audio.mp3", format="mp3")  #("/content/gdrive/My Drive/AI/audio1.mp3", format="mp3")
+    print('flac audio converted to mp3 audio' )
     
     # loading video dsa gfg intro video
     videoclip = VideoFileClip('my_vid_tmp.mp4') #("/content/gdrive/My Drive/AI/my_video1.mp4")
-   
+    print('video clip loaded in first time')   
     # loading audio file
     audioclip = AudioFileClip('audio.mp3') #.subclip(0, 15)
-  
+    print('mp3 format audio clip loaded in')  
     # adding audio to the video clip
     mergedclip = videoclip.set_audio(audioclip)
-  
+    print('video and audio merged')  
     # showing video clip
     #videoclip.ipython_display()
     
@@ -98,9 +102,12 @@ def engine(text_input):
      
     #Convert text to speech using facebook's latest model from HF hub   
     speech = text2speech(text_input)
-     
+    
+    #getting audio clip's duration
+    audio_length = int(WAVE("alarm.wav").info.length)
+      
     #Calculate the desired frame per second based on given audio length and entities identified
-    fps= entities_num / 19 #length of audio file   #13 / 19
+    fps= entities_num / audio_length #19 #length of audio file   #13 / 19
     fps = float(format(fps, '.5f'))
     print('fps is: ',fps)