Update app.py

app.py

@@ -10,15 +10,15 @@ from sklearn.feature_extraction.text import TfidfVectorizer
 from nltk.tokenize import sent_tokenize
 from nltk.corpus import stopwords
 
-# Download NLTK data once
+# 📦 Download NLTK data (ideally run this in setup instead of here)
 nltk.download('punkt_tab')
 nltk.download('stopwords')
 
-# Global objects (faster)
-stop_words = set(stopwords.words('english'))
-summarizer = pipeline("summarization", model="sshleifer/distilbart-cnn-12-6", device=-1)  # Use device=0 if GPU
+# 🔄 Global models (load once)
+whisper_model = whisper.load_model("tiny.en")  # Use 'base.en' for better accuracy if on GPU
+summarizer = pipeline("summarization", model="t5-small", device=-1)  # Use device=0 if GPU available
 
-# 🧠 Function to download YouTube video
+# 🔽 Download YouTube video
 def download_youtube_video(youtube_url, filename="youtube_video.mp4"):
     print(f"⬇️ Downloading YouTube video via yt-dlp: {youtube_url}")
     command = ["yt-dlp", "-f", "best[ext=mp4]+bestaudio/best", "-o", filename, youtube_url]
@@ -27,36 +27,29 @@ def download_youtube_video(youtube_url, filename="youtube_video.mp4"):
         raise Exception("YouTube download failed: " + result.stderr)
     return filename
 
-# 🎧 Audio extraction
+# 🎧 Extract audio from video
 def extract_audio(video_path):
     clip = VideoFileClip(video_path)
     audio_path = "temp_audio.wav"
-    clip.audio.write_audiofile(audio_path, codec='pcm_s16le')
+    clip.audio.write_audiofile(audio_path, codec='pcm_s16le', verbose=False, logger=None)
     return audio_path
 
-# 📝 Audio transcription
+# 📝 Transcribe audio using Whisper
 def transcribe_audio(audio_path):
-    model = whisper.load_model("tiny.en")  # Use "base.en" for slightly better quality
-    result = model.transcribe(audio_path)
+    result = whisper_model.transcribe(audio_path)
     return result["text"]
 
-# 📄 Summary generator with batching
+# 📄 Generate summary in chunks
 def generate_summary(text, default_max_len=130, default_min_len=30):
-    summarizer = pipeline("summarization")
     sentences = sent_tokenize(text)
-
-    # Chunk sentences in groups of 10
     chunks = [' '.join(sentences[i:i + 10]) for i in range(0, len(sentences), 10)]
     summary = ""
 
     for chunk in chunks:
         input_len = len(chunk.split())
-
-        # Dynamically scale max and min length
         dynamic_max = max(20, min(default_max_len, input_len - 1))
         dynamic_min = max(10, min(default_min_len, dynamic_max - 10))
 
-        # Avoid max_length > input_length error
         result = summarizer(
             chunk,
             max_length=dynamic_max,
@@ -68,12 +61,11 @@ def generate_summary(text, default_max_len=130, default_min_len=30):
 
     return summary.strip()
 
-# ❓ Quiz generator
+# ❓ Generate quiz
 def generate_quiz(text, num_questions=5):
     sentences = sent_tokenize(text)
     tfidf = TfidfVectorizer(stop_words='english', max_features=300)
     X = tfidf.fit_transform(sentences)
-
     quiz = []
     used = set()
 
@@ -100,7 +92,7 @@ def generate_quiz(text, num_questions=5):
         for i, q in enumerate(quiz)
     ])
 
-# 📺 Subtitle formatter
+# 📺 Subtitle formatting
 def generate_subtitles(text, max_words_per_line=10):
     sentences = sent_tokenize(text)
     subtitles = []
@@ -112,15 +104,14 @@ def generate_subtitles(text, max_words_per_line=10):
             count += 1
     return "\n".join(subtitles)
 
-# 🧪 Master function
+# 🧪 Main processor
 def process_video(video_path, selected_services):
     results = {}
     print("🔧 Extracting audio...")
     audio_path = extract_audio(video_path)
 
-    transcription = transcribe_audio(audio_path) if "Transcription" in selected_services else None
-
-    if transcription:
+    if "Transcription" in selected_services:
+        transcription = transcribe_audio(audio_path)
         results["transcription"] = transcription
 
         if "Summary" in selected_services: