1

app.py

@@ -33,30 +33,35 @@ def resize(img_list):
 #               for img in resize_img_list
 #               if img.endswith(".jpg")]
 #print(image_files)
-def images_to_video(fps, resize_img_list):
-    clip = moviepy.video.io.ImageSequenceClip.ImageSequenceClip(resize_img_list, fps=fps)
-    return clip
+
+#def images_to_video(fps, resize_img_list):
+#    clip = moviepy.video.io.ImageSequenceClip.ImageSequenceClip(resize_img_list, fps=fps)
+#    return clip
 #clip.write_videofile('/content/gdrive/My Drive/AI/my_vid_20apr.mp4')
 
 
-def merge_audio_video(speech, clip):
+def merge_audio_video(fps, resize_img_list, speech):
+    #String a list of images into a video and write to memory
+    clip = moviepy.video.io.ImageSequenceClip.ImageSequenceClip(resize_img_list, fps=fps)
+    clip.write_videofile('my_vid_tmp.mp4')
+    
     #convert flac to mp3 audio format 
     wav_audio = AudioSegment.from_file(speech, "flac")  #("/content/gdrive/My Drive/AI/audio1.flac", "flac")
     wav_audio.export("audio.mp3", format="mp3")  #("/content/gdrive/My Drive/AI/audio1.mp3", format="mp3")
     
     # loading video dsa gfg intro video
-    clip = VideoFileClip(clip) #("/content/gdrive/My Drive/AI/my_video1.mp4")
+    videoclip = VideoFileClip('my_vid_tmp.mp4') #("/content/gdrive/My Drive/AI/my_video1.mp4")
    
     # loading audio file
     audioclip = AudioFileClip('audio.mp3') #.subclip(0, 15)
   
     # adding audio to the video clip
-    videoclip = clip.set_audio(audioclip)
+    mergedclip = videoclip.set_audio(audioclip)
   
     # showing video clip
     #videoclip.ipython_display()
     
-    return videoclip
+    return mergedclip
     
 
 fastspeech = gr.Interface.load("huggingface/facebook/fastspeech2-en-ljspeech")
@@ -88,10 +93,10 @@ def engine(text_input):
     fps= entities_num / 19 #length of audio file   #13 / 19
      
     #Convert string of images into a video
-    clip = images_to_video(fps, resize_img_list)
+    #clip = images_to_video(fps, resize_img_list)
      
     #Merge video and audio created above
-    merged_file = merge_audio_video(speech, clip)
+    mergedclip = merge_audio_video(fps, resize_img_list, speech)
     #{'prompt':text_input,'steps':'50','width':'256','height':'256','images':'1','scale':10}).launch()
     #img_intfc = gr.Interface.load("spaces/multimodalart/latentdiffusion", inputs=[gr.inputs.Textbox(lines=1, label="Input Text"), gr.inputs.Textbox(lines=1, label="Input Text"), gr.inputs.Textbox(lines=1, label="Input Text"), gr.inputs.Textbox(lines=1, label="Input Text"), gr.inputs.Textbox(lines=1, label="Input Text"), gr.inputs.Textbox(lines=1, label="Input Text")],
     #outputs=[gr.outputs.Image(type="pil", label="output image"),gr.outputs.Carousel(label="Individual images",components=["image"]),gr.outputs.Textbox(label="Error")], )
@@ -108,7 +113,7 @@ def engine(text_input):
     #run(prompt, steps, width, height, images, scale)
     
     
-    return merged_file #img, entities, speech
+    return mergedclip #img, entities, speech
     
 #image = gr.outputs.Image(type="pil", label="output image")
 app = gr.Interface(engine,