Update audio extraction libraries

app.py

@@ -11,6 +11,7 @@ from functools import reduce
 from sklearn.preprocessing import StandardScaler
 from matplotlib import pyplot as plt
 import tempfile
+import yt_dlp
 
 # Constants
 SR = 12000
@@ -26,6 +27,7 @@ tf.get_logger().setLevel('ERROR')
 
 def extract_audio(url):
     try:
+        # Attempt to use pytube
         yt = YouTube(url)
         video_title = yt.title
         audio_stream = yt.streams.filter(only_audio=True).first()
@@ -42,7 +44,28 @@ def extract_audio(url):
             st.error("No audio stream found")
             return None, None, None
     except Exception as e:
-        st.error(f"An error occurred: {e}")
+        st.warning(f"pytube error: {e}. Falling back to yt-dlp.")
+        return extract_audio_with_ytdlp(url)
+
+def extract_audio_with_ytdlp(url):
+    try:
+        ydl_opts = {
+            'format': 'bestaudio/best',
+            'postprocessors': [{
+                'key': 'FFmpegExtractAudio',
+                'preferredcodec': 'mp3',
+                'preferredquality': '192',
+            }],
+            'outtmpl': '%(title)s.%(ext)s',
+            'paths': {'home': tempfile.mkdtemp()}
+        }
+        with yt_dlp.YoutubeDL(ydl_opts) as ydl:
+            info_dict = ydl.extract_info(url, download=True)
+            video_title = info_dict.get('title', None)
+            audio_file = os.path.join(ydl_opts['paths']['home'], f"{video_title}.mp3")
+            return audio_file, video_title, ydl_opts['paths']['home']
+    except Exception as e:
+        st.error(f"An error occurred with yt-dlp: {e}")
         return None, None, None