Update app.py

app.py

@@ -13,12 +13,16 @@ from bs4 import SoupStrainer
 from youtube_transcript_api import YouTubeTranscriptApi, TranscriptsDisabled, NoTranscriptFound
 import yt_dlp
 import re
+from googleapiclient.discovery import build
+from googleapiclient.errors import HttpError
 
 # Load environment variables (optional)
 load_dotenv()
 
 # Hardcoded Groq API key 
 GROQ_API_KEY = "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"
+# YouTube API key (to be set in Hugging Face Spaces secrets)
+YOUTUBE_API_KEY = os.getenv("YOUTUBE_API_KEY")
 
 # Custom CSS
 st.markdown("""
@@ -123,7 +127,7 @@ if "llm" not in st.session_state:
     st.session_state.llm = ChatGroq(
         api_key=GROQ_API_KEY,
         model="llama3-70b-8192",
-        max_tokens=512  # Further reduced to minimize resource usage
+        max_tokens=512  # Keep reduced to minimize resource usage
     )
 
 # Sidebar for URL and YouTube input
@@ -203,6 +207,45 @@ def fetch_youtube_transcript(video_id):
         st.error(f"Error fetching transcript with youtube-transcript-api: {str(e)}")
         return None
 
+# Function to fetch captions using YouTube Data API (limited to listing with API key)
+def fetch_youtube_captions_api(video_id, api_key):
+    if not api_key:
+        st.warning("YOUTUBE_API_KEY not set. Skipping YouTube Data API fallback.")
+        return None
+    try:
+        youtube = build('youtube', 'v3', developerKey=api_key)
+        captions = youtube.captions().list(
+            part='snippet',
+            videoId=video_id
+        ).execute()
+
+        caption_id = None
+        for item in captions.get('items', []):
+            if item['snippet']['language'] == 'en':
+                caption_id = item['id']
+                break
+            elif item['snippet']['language'] in ['en-US', 'en-GB']:
+                caption_id = item['id']
+                break
+
+        if not caption_id:
+            st.warning("No English captions found via YouTube Data API.")
+            return None
+
+        # Note: Downloading captions requires OAuth 2.0 authentication
+        st.warning(
+            "English captions are available for this video but cannot be fetched with an API key alone. "
+            "Downloading captions requires OAuth 2.0 authentication, which is not supported in Hugging Face Spaces without user interaction. "
+            "To fetch captions:\n"
+            "- Test locally with OAuth 2.0 setup (see https://developers.google.com/youtube/v3/guides/auth/installed-apps for instructions).\n"
+            "- Or try a video with transcripts available (e.g., https://www.youtube.com/watch?v=dQw4w9WgXcQ)."
+        )
+        return None
+
+    except HttpError as e:
+        st.error(f"Error fetching captions with YouTube Data API: {str(e)}")
+        return None
+
 # Function to extract subtitles using yt-dlp with cookies
 def extract_subtitles_with_ytdlp(video_url):
     ydl_opts = {
@@ -360,6 +403,10 @@ if process_youtube_clicked:
                         st.text("Fetching Closed Captions...Started...✅✅✅")
                         transcript_text = extract_subtitles_with_ytdlp(youtube_url)
 
+                        if not transcript_text and YOUTUBE_API_KEY:
+                            st.text("Fetching Captions via YouTube Data API...Started...✅✅✅")
+                            transcript_text = fetch_youtube_captions_api(video_id, YOUTUBE_API_KEY)
+
                         if not transcript_text:
                             st.error(
                                 "No transcripts or closed captions available. "
@@ -369,6 +416,7 @@ if process_youtube_clicked:
                                 "Solutions:\n"
                                 "- Ensure captions are enabled for the video by checking the video settings on YouTube (gear icon > Subtitles/CC > Enable if available).\n"
                                 "- Regenerate and upload a fresh cookies.txt file (see instructions above).\n"
+                                "- Ensure YOUTUBE_API_KEY is set in Spaces secrets (Settings > Secrets > Add YOUTUBE_API_KEY).\n"
                                 "- Try a different video (e.g., https://www.youtube.com/watch?v=dQw4w9WgXcQ, which has transcripts available).\n"
                                 "- Test locally to rule out Hugging Face Spaces IP restrictions by running: pip install -r requirements.txt && streamlit run app.py"
                             )