Update app.py

app.py

@@ -1,36 +1,43 @@
 import gradio as gr
-from utils import generate_script, generate_audio, truncate_text
+from utils import generate_script, generate_audio, truncate_text, extract_text_from_url
 from prompts import SYSTEM_PROMPT
 from pydub import AudioSegment
 import pypdf
 import os
 import tempfile
 
-def generate_podcast(file, tone, length):
-    # Extract text from PDF
-    if not file.name.lower().endswith('.pdf'):
-        raise gr.Error("Please upload a PDF file.")
+def generate_podcast(file, url, tone, length):
+    if file and url:
+        raise gr.Error("Please provide either a PDF file or a URL, not both.")
     
-    try:
-        pdf_reader = pypdf.PdfReader(file.name)
-        text = ""
-        for page in pdf_reader.pages:
-            text += page.extract_text()
-    except Exception as e:
-        raise gr.Error(f"Error reading the PDF file: {str(e)}")
+    if file:
+        if not file.name.lower().endswith('.pdf'):
+            raise gr.Error("Please upload a PDF file.")
+        
+        try:
+            pdf_reader = pypdf.PdfReader(file.name)
+            text = ""
+            for page in pdf_reader.pages:
+                text += page.extract_text()
+        except Exception as e:
+            raise gr.Error(f"Error reading the PDF file: {str(e)}")
+    elif url:
+        try:
+            text = extract_text_from_url(url)
+        except Exception as e:
+            raise gr.Error(f"Error extracting text from URL: {str(e)}")
+    else:
+        raise gr.Error("Please provide either a PDF file or a URL.")
     
-    # Truncate text to 2048 tokens
     truncated_text = truncate_text(text)
     if len(truncated_text) < len(text):
         print("Warning: The input text was truncated to fit within 2048 tokens.")
     
-    # Generate script
     try:
         script = generate_script(SYSTEM_PROMPT, truncated_text, tone, length)
     except Exception as e:
         raise gr.Error(f"Error generating script: {str(e)}")
     
-    # Generate audio for each dialogue item
     audio_segments = []
     transcript = ""
     try:
@@ -43,30 +50,27 @@ def generate_podcast(file, tone, length):
     except Exception as e:
         raise gr.Error(f"Error generating audio: {str(e)}")
     
-    # Combine audio segments
     combined_audio = sum(audio_segments)
     
-    # Save combined audio to a temporary file
     with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as temp_audio:
         combined_audio.export(temp_audio.name, format="mp3")
         temp_audio_path = temp_audio.name
     
     return temp_audio_path, transcript
 
-# Gradio interface setup
 instructions = """
 # Podcast Generator
 
 Welcome to the Podcast Generator project! This tool allows you to create custom podcast episodes using AI-generated content.
 
 ## Features
-* Generate podcast scripts from PDF content
+* Generate podcast scripts from PDF content or web pages
 * Convert text to speech for a natural listening experience
 * Choose the tone of your podcast
 * Export episodes as MP3 files
 
 ## How to Use
-1. Upload a PDF file (content will be truncated to 2048 tokens if longer)
+1. Upload a PDF file OR enter a URL (content will be truncated to 2048 tokens if longer)
 2. Select the desired tone (humorous, casual, formal)
 3. Choose the podcast length
 4. Click "Generate" to create your podcast
@@ -78,7 +82,8 @@ Note: This tool uses the LLaMa 3.1 70B model for script generation and gTTS for
 iface = gr.Interface(
     fn=generate_podcast,
     inputs=[
-        gr.File(label="Upload PDF file", file_types=[".pdf"]),
+        gr.File(label="Upload PDF file (optional)", file_types=[".pdf"]),
+        gr.Textbox(label="OR Enter URL"),
         gr.Radio(["humorous", "casual", "formal"], label="Select podcast tone", value="casual"),
         gr.Radio(["Short (1-2 min)", "Medium (3-5 min)"], label="Podcast length", value="Medium (3-5 min)")
     ],