Update app.py

app.py

@@ -1,65 +1,92 @@
-import gradio as gr 
-import os
-import sys
-import subprocess
-#from moviepy.editor import VideoFileClip
-
+import gradio as gr
 import whisper
 from whisper.utils import write_vtt
 
-model = whisper.load_model("medium")
-
-title = "Add Text/Caption to your YouTube Shorts - MultiLingual"
-
-def video2mp3(video_file, output_ext="mp3"):
-    filename, ext = os.path.splitext(video_file)
-    subprocess.call(["ffmpeg", "-y", "-i", video_file, f"{filename}.{output_ext}"], 
-                    stdout=subprocess.DEVNULL,
-                    stderr=subprocess.STDOUT)
-    return f"{filename}.{output_ext}"
-
-
-def translate(input_video):
-
-    audio_file = video2mp3(input_video)
-    
-    options = dict(beam_size=5, best_of=5, fp16 = False)
-    translate_options = dict(task="translate", **options)
-    result = model.transcribe(audio_file,**translate_options)
-
-    output_dir = ''
-    audio_path = audio_file.split(".")[0]
-
-    with open(os.path.join(output_dir, audio_path + ".vtt"), "w") as vtt:
-      write_vtt(result["segments"], file=vtt)
-
-    subtitle = audio_path + ".vtt"
-    output_video = audio_path + "_subtitled.mp4"
-
-    os.system(f"ffmpeg -i {input_video} -vf subtitles={subtitle} {output_video}")
-
-    return output_video
+from pytube import YouTube
+import os
+import sys
+import subprocess
 
+loaded_model = whisper.load_model("base")
+current_size = 'base'
+
+def inference(link):
+
+  yt = YouTube(link)
+  
+  audio_path = yt.streams.filter(only_audio=True)[0].download(filename="audio.mp4")
+  print(f'audio path :  {audio_path}')
+  video_path = yt.streams.filter(file_extension='mp4')
+  
+  #options = whisper.DecodingOptions(without_timestamps=True)
+  options = dict(beam_size=5, best_of=5, fp16 = False)
+  translate_options = dict(task="inference", **options)
+  results = loaded_model.transcribe(audio_path,**translate_options)
+  
+  #output_dir = ''
+  path = audio_path.split(".")[0]
+  print(path)
+
+  with open(path + ".vtt", "w") as vtt:
+    write_vtt(results["segments"], file=vtt)  
+
+  subtitle = path + ".vtt"
+  output_video = path + "_subtitled.mp4"
+
+  os.system(f"ffmpeg -i {video_path} -vf subtitles={subtitle} {output_video}")
+  
+  return output_video
+  
+def change_model(size):
+  if size == current_size:
+    return
+  loaded_model = whisper.load_model(size)
+  current_size = size
+
+def populate_metadata(link):
+  yt = YouTube(link)
+  return yt.thumbnail_url, yt.title
+
+title="Youtube Caption Generator"
+description="Generate captions of Youtube videos using OpenAI's Whisper"
 block = gr.Blocks()
-with block:
 
+with block:
+    gr.HTML(
+        """
+            <div style="text-align: center; max-width: 500px; margin: 0 auto;">
+              <div>
+                <h1>Youtube Caption Generator</h1>
+              </div>
+              <p style="margin-bottom: 10px; font-size: 94%">
+                Generate captions of Youtube videos using OpenAI's Whisper
+              </p>
+            </div>
+        """
+    )
     with gr.Group():
         with gr.Box():
-            with gr.Row().style():
-                inp_video = gr.Video(
-                    label="Input Video",
-                    type="filepath",
-                    mirror_webcam = False
-                )
-                op_video = gr.Video()
-            btn = gr.Button("Generate Subtitle Video")
-        
-        
-        
-
-
-        
-        btn.click(translate, inputs=[inp_video], outputs=[op_video])
-        
-        
-block.launch(enable_queue = True)
+          sz = gr.Dropdown(label="Model Size", choices=['base','small', 'medium', 'large'], value='base')
+          
+          link = gr.Textbox(label="YouTube Link")
+          
+          with gr.Row().style(mobile_collapse=False, equal_height=True):
+            title = gr.Label(label="Video Title", placeholder="Title")
+            img = gr.Image(label="Thumbnail")
+            
+        #   text = gr.Textbox(
+        #       label="Transcription", 
+        #       placeholder="Transcription Output",
+        #       lines=5)
+          
+          op_video = gr.Video()
+
+          with gr.Row().style(mobile_collapse=False, equal_height=True): 
+              btn = gr.Button("Generate Captions")       
+          
+          # Events
+          btn.click(inference, inputs=[link], outputs=[op_video])
+          link.change(populate_metadata, inputs=[link], outputs=[img, title])
+          sz.change(change_model, inputs=[sz], outputs=[])
+
+block.launch(debug=True,enable_queue=True)