Create app.py

app.py

@@ -0,0 +1,60 @@
+import gradio as gr
+import whisper
+import yt_dlp
+import os
+from transformers import pipeline
+
+# Load Whisper model
+whisper_model = whisper.load_model("base")
+
+# Load Summarization pipeline (you can replace with a LLaMA fine-tuned model if you have one)
+summarizer = pipeline("summarization", model="facebook/bart-large-cnn")
+
+def download_youtube_audio(youtube_url):
+    filename = "audio.mp3"
+    ydl_opts = {
+        'format': 'bestaudio/best',
+        'outtmpl': filename,
+        'quiet': True,
+        'postprocessors': [{
+            'key': 'FFmpegExtractAudio',
+            'preferredcodec': 'mp3',
+            'preferredquality': '192',
+        }],
+    }
+
+    with yt_dlp.YoutubeDL(ydl_opts) as ydl:
+        ydl.download([youtube_url])
+
+    return filename
+
+def transcribe_and_summarize(youtube_url):
+    try:
+        # Step 1: Download audio
+        audio_file = download_youtube_audio(youtube_url)
+
+        # Step 2: Transcribe
+        result = whisper_model.transcribe(audio_file)
+        transcript = result["text"]
+
+        # Step 3: Summarize
+        summary = summarizer(transcript, max_length=200, min_length=50, do_sample=False)[0]["summary_text"]
+
+        # Clean up
+        os.remove(audio_file)
+
+        return f"📄 **Transcript:**\n\n{transcript}\n\n🧠 **Summary:**\n\n{summary}"
+
+    except Exception as e:
+        return f"❌ Error: {str(e)}"
+
+# Gradio UI
+demo = gr.Interface(
+    fn=transcribe_and_summarize,
+    inputs=gr.Textbox(label="Enter YouTube Video URL"),
+    outputs=gr.Markdown(label="Transcript and Summary"),
+    title="🎥 YouTube Video Summarizer",
+    description="Enter a YouTube video link. This app will transcribe it and summarize using AI."
+)
+
+demo.launch()