Update app.py

app.py

@@ -126,7 +126,7 @@ with st.sidebar:
     process_url_clicked = st.button("Process URL")
 
     st.header("Enter YouTube URL")
-    youtube_url = st.text_input("YouTube URL", placeholder="e.g., https://www.youtube.com/watch?v=dQw4w9WgXcQ")
+    youtube_url = st.text_input("YouTube URL", placeholder="e.g., https://www.youtube.com/watch?v=pxiP-HJLCx0")
     process_youtube_clicked = st.button("Process YouTube Video")
 
 # Main content container
@@ -240,11 +240,34 @@ if process_youtube_clicked:
                         st.stop()
 
                     st.text("Fetching Transcript...Started...✅✅✅")
-                    transcript = YouTubeTranscriptApi.get_transcript(video_id, languages=['en', 'bn'])
-                    transcript_text = " ".join([entry['text'] for entry in transcript])
-                    
+                    # Get the list of available transcripts
+                    transcript_list = YouTubeTranscriptApi.list_transcripts(video_id)
+
+                    transcript = None
+                    # Try to find a manually created transcript in English
+                    try:
+                        transcript = transcript_list.find_manually_created_transcript(['en'])
+                    except NoTranscriptFound:
+                        # If no manual transcript, try an auto-generated one in English
+                        try:
+                            transcript = transcript_list.find_generated_transcript(['en'])
+                        except NoTranscriptFound:
+                            # If no English transcript, get the first available transcript and translate to English
+                            for t in transcript_list:
+                                if t.is_translatable:
+                                    transcript = t.translate('en')
+                                    break
+
+                    if not transcript:
+                        st.error("No transcript available or translatable to English. Try a different video.")
+                        st.stop()
+
+                    # Fetch the transcript data
+                    transcript_data = transcript.fetch()
+                    transcript_text = " ".join([entry['text'] for entry in transcript_data])
+
                     if not transcript_text.strip():
-                        st.error("No transcript available for this video. Try a different video.")
+                        st.error("Transcript is empty. Try a different video.")
                         st.stop()
 
                     # Create a Document object from the transcript
@@ -253,10 +276,10 @@ if process_youtube_clicked:
                     embeddings = st.session_state.embeddings
                     process_content([doc], embeddings)
                 except TranscriptsDisabled:
-                    st.error("Transcripts are disabled for this video. Try a different video.")
+                    st.error("Transcripts are disabled for this video by the creator. Try a different video.")
                     st.stop()
                 except NoTranscriptFound:
-                    st.error("No transcript found in the supported languages (English or Bengali). Try a different video.")
+                    st.error("No transcript found in any language. Try a different video.")
                     st.stop()
                 except Exception as e:
                     st.error(f"Error processing YouTube video: {str(e)}")