Update app.py

app.py

@@ -7,12 +7,37 @@ def transcribe(audio):
     text = pipe(audio)["text"]
     return text
 
-iface = gr.Interface(
+def transcribe_video(link):
+    video=link
+    data=pytube.YouTube(video)
+    audio=data.streams.get_audio_only()
+    audio.download()
+
+    pattern=r".mp4$"
+    content=os.listdir()
+    for i in content:
+        if re.search(pattern,i) is not None:
+            video=i
+    
+    text = pipe(video)["text"]
+    return text
+
+iface1 = gr.Interface(
     fn=transcribe, 
     inputs=gr.Audio(source="microphone", type="filepath"), 
     outputs="text",
     title="Whisper Base",
-    description="Realtime demo for  speech recognition using a fine-tuned Whisper base model.",
+    description="Realtime demo for  speech recognition using a fine-tuned Whisper-base model.",
 )
 
-iface.launch()
+iface2 = gr.Interface(
+    fn=transcribe, 
+    inputs=gr.Textbox(label="Youtube Link",placeholder="Youtube Link"), 
+    outputs=["text"],
+    title="Whisper Base",
+    description="Asynchronous demo for  youtube speech recognition using a fine-tuned Whisper-base model.",
+)
+
+
+
+gr.Parallel(iface1,iface2).launch()