Update app.py

app.py

@@ -24,6 +24,11 @@ except ImportError as e:
     logger.error("Please ensure moviepy is installed correctly")
     raise
 
+# Supported file formats
+AUDIO_FORMATS = ['.wav', '.mp3', '.ogg', '.flac', '.aac', '.m4a', '.wma']
+VIDEO_FORMATS = ['.mp4', '.avi', '.mov', '.flv', '.wmv', '.mkv', '.webm']
+SUPPORTED_FORMATS = AUDIO_FORMATS + VIDEO_FORMATS
+
 # Initialize the Dash app
 app = Dash(__name__, external_stylesheets=[dbc.themes.BOOTSTRAP])
 
@@ -72,38 +77,46 @@ app.layout = dbc.Container([
 
 def process_media(file_path, is_url=False):
     global generated_file, transcription_text
-    temp_audio_file = None
+    temp_file = None
+    wav_path = None
     try:
         if is_url:
             if 'youtube.com' in file_path or 'youtu.be' in file_path:
                 yt = YouTube(file_path)
-                stream = yt.streams.filter(only_audio=True).first()
-                temp_audio_file = tempfile.NamedTemporaryFile(delete=False, suffix='.mp4')
-                stream.download(output_path=os.path.dirname(temp_audio_file.name), filename=os.path.basename(temp_audio_file.name))
+                stream = yt.streams.filter(progressive=True, file_extension='mp4').order_by('resolution').desc().first()
+                temp_file = tempfile.NamedTemporaryFile(delete=False, suffix='.mp4')
+                stream.download(output_path=os.path.dirname(temp_file.name), filename=os.path.basename(temp_file.name))
             else:
                 response = requests.get(file_path)
-                temp_audio_file = tempfile.NamedTemporaryFile(delete=False)
-                temp_audio_file.write(response.content)
-                temp_audio_file.close()
+                content_type = response.headers.get('content-type', '')
+                if 'audio' in content_type:
+                    suffix = '.mp3'
+                elif 'video' in content_type:
+                    suffix = '.mp4'
+                else:
+                    suffix = ''
+                temp_file = tempfile.NamedTemporaryFile(delete=False, suffix=suffix)
+                temp_file.write(response.content)
+                temp_file.close()
         else:
-            temp_audio_file = tempfile.NamedTemporaryFile(delete=False)
-            temp_audio_file.write(file_path)
-            temp_audio_file.close()
+            temp_file = tempfile.NamedTemporaryFile(delete=False)
+            temp_file.write(file_path)
+            temp_file.close()
 
-        file_extension = os.path.splitext(temp_audio_file.name)[1].lower()
+        file_extension = os.path.splitext(temp_file.name)[1].lower()
         
-        if file_extension in ['.mp4', '.avi', '.mov', '.flv', '.wmv']:
-            video = VideoFileClip(temp_audio_file.name)
+        if file_extension in VIDEO_FORMATS:
+            video = VideoFileClip(temp_file.name)
             audio = video.audio
-            wav_path = temp_audio_file.name + ".wav"
+            wav_path = temp_file.name + ".wav"
             audio.write_audiofile(wav_path)
             video.close()
-        elif file_extension in ['.wav', '.mp3', '.ogg', '.flac']:
-            audio = AudioSegment.from_file(temp_audio_file.name)
-            wav_path = temp_audio_file.name + ".wav"
+        elif file_extension in AUDIO_FORMATS:
+            audio = AudioSegment.from_file(temp_file.name)
+            wav_path = temp_file.name + ".wav"
             audio.export(wav_path, format="wav")
         else:
-            return "Unsupported file format. Please upload an audio or video file.", False
+            return f"Unsupported file format: {file_extension}. Please upload a supported audio or video file.", False
 
         with open(wav_path, "rb") as audio_file:
             transcript = openai.Audio.transcribe("whisper-1", audio_file)
@@ -126,11 +139,11 @@ def process_media(file_path, is_url=False):
         logger.error(f"Error during processing: {str(e)}")
         return f"An error occurred: {str(e)}", False
     finally:
-        if temp_audio_file and os.path.exists(temp_audio_file.name):
-            os.unlink(temp_audio_file.name)
-        if 'wav_path' in locals() and os.path.exists(wav_path):
+        if temp_file and os.path.exists(temp_file.name):
+            os.unlink(temp_file.name)
+        if wav_path and os.path.exists(wav_path):
             os.unlink(wav_path)
-
+            
 @app.callback(
     [Output('output-media-upload', 'children'),
      Output('transcription-status', 'children'),