add intro

README.md

@@ -1,4 +1,7 @@
 # Webpage To Podcast
+
+git add remote https://huggingface.co/spaces/vincentclaes/webpage-to-podcast
+
 Take any webpage and create a podcast out of it
 
 ## Prompt

app.py

@@ -1,16 +1,38 @@
-# import numpy as np
-import gradio
+import gradio as gr
 from openai import OpenAI
 import tempfile
-import os
+import time
+import typing as tp
+from pathlib import Path
+from pydub import AudioSegment
 
 from dotenv import load_dotenv
 load_dotenv()
 
-import uuid
+client = OpenAI()
 
+class FileCleaner:
+    def __init__(self, file_lifetime: float = 3600):
+        self.file_lifetime = file_lifetime
+        self.files = []
 
-client = OpenAI()
+    def add(self, path: tp.Union[str, Path]):
+        self._cleanup()
+        self.files.append((time.time(), Path(path)))
+
+    def _cleanup(self):
+        now = time.time()
+        for time_added, path in list(self.files):
+            if now - time_added > self.file_lifetime:
+                if path.exists():
+                    path.unlink()
+                self.files.pop(0)
+            else:
+                break
+
+
+file_cleaner = FileCleaner()
+intro = AudioSegment.from_mp3("intro.mp3")
 
 DEFAULT_SYSTEM_PROMPT = """
 You are a podcast editor that specialized to create a script out of a webpage.
@@ -33,13 +55,6 @@ You are a podcast editor that specialized to create a script out of a webpage.
 """
 
 def generate_episode(system_prompt, weblink):
-    # sr = 48000
-    # a4_freq, tones_from_a4 = 440, 12 * (octave - 4) + (note - 9)
-    # frequency = a4_freq * 2 ** (tones_from_a4 / 12)
-    # duration = int(duration)
-    # audio = np.linspace(0, duration, duration * sr)
-    # audio = (20000 * np.sin(audio * (2 * np.pi * frequency))).astype(np.int16)
-    # return sr, audio
     response = client.chat.completions.create(
         model="gpt-4o",
         messages=[
@@ -54,34 +69,34 @@ def generate_episode(system_prompt, weblink):
         input=script,
     )
         
-    with tempfile.NamedTemporaryFile(suffix=".mp3", delete=False) as temp_file:
-        temp_file.write(response.content)    
-    # state["mp3_file"] = temp_file.name
-    return temp_file.name
+    with tempfile.NamedTemporaryFile(suffix=".mp3", delete=False) as podcast_file:
+        podcast_file.write(response.content) 
+        file_cleaner.add(podcast_file.name)
 
+    podcast = AudioSegment.from_mp3(podcast_file.name)
+    merged_audio = intro + podcast
+    
+    with tempfile.NamedTemporaryFile(suffix=".mp3", delete=False) as final_file:
+        merged_audio.export(final_file.name, format="mp3")
 
-def cleanup(state):
-    mp3_file_name = state.get("mp3_file")    
-    if mp3_file_name and os.path.exists(mp3_file_name):
-        os.remove(mp3_file_name)
-        print(f"Removed file: {mp3_file_name}")
-    else:
-        print(f"No file found to delete: {mp3_file_name}")
+    print("MP3 files merged successfully!")
 
-demo = gradio.Interface(
+    return final_file.name, script
+
+demo = gr.Interface(
     fn=generate_episode,
     inputs=[
-        gradio.Textbox(value=DEFAULT_SYSTEM_PROMPT, label="System Prompt"),
-        gradio.Textbox(value="https://en.wikipedia.org/wiki/Mount_Tambora", label="Weblink"),
-        # gradio.State(),  # State to track the mp3 file
-
+        gr.Textbox(label="System Prompt"),
+        gr.Textbox(label="Weblink"),
+    ],
+    outputs=[gr.Audio(label="Podcast Audio"), gr.Textbox(label="Podcast Script")],
+    examples=[
+        [DEFAULT_SYSTEM_PROMPT, "https://en.wikipedia.org/wiki/Mount_Tambora"],
+        [DEFAULT_SYSTEM_PROMPT, "https://en.wikipedia.org/wiki/Great_Wall_of_China"],
+        [DEFAULT_SYSTEM_PROMPT, "https://en.wikipedia.org/wiki/Apollo_11"],
     ],
-    # outputs=["audio", gradio.State()],
-    outputs="audio",
-
 )
 
-# demo.cleanup(cleanup)
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()

requirements.txt

@@ -2,3 +2,4 @@ gradio
 numpy
 python-dotenv
 openai
+pydub