Refactor transcribe function to handle both video

app.py

@@ -10,35 +10,37 @@ import os
 def transcribe_audio(audio_file):
     text = "Test text"
     # text = pipe(audio_file)["text"]
-    os.remove(audio_file)  # Remove temporary audio file
     return text
 
 
-def transcribe(input_data):
-    if input_data["audio"]:
-        return transcribe_audio(input_data["audio"].name)
-    elif input_data["video_url"]:
-        video_url = input_data["video_url"]
-        # Download the video from the URL
-        video_filename = "temp_video.mp4"
-        with open(video_filename, "wb") as f:
-            response = requests.get(video_url)
-            f.write(response.content)
+def transcribe_video(video_url):
+    # Download the video from the URL
+    video_filename = "temp_video.mp4"
+    with open(video_filename, "wb") as f:
+        response = requests.get(video_url)
+        f.write(response.content)
 
-        # Load the video using moviepy
-        video = VideoFileClip(video_filename)
-        audio = video.audio
+    # Load the video using moviepy
+    video = VideoFileClip(video_filename)
+    audio = video.audio
 
-        audio_file = "temp_audio.wav"
-        audio.write_audiofile(audio_file, codec="pcm_s16le")
+    audio_file = "temp_audio.wav"
+    audio.write_audiofile(audio_file, codec="pcm_s16le")
 
-        text = transcribe_audio(audio_file)
+    text = transcribe_audio(audio_file)
 
-        # Remove temporary files
-        os.remove(video_filename)
-        os.remove(audio_file)
+    # Remove temporary files
+    os.remove(video_filename)
+    os.remove(audio_file)
 
-        return text
+    return text
+
+
+def transcribe(video_url="", audio=None):
+    if audio:
+        return transcribe_audio(audio)
+    elif video_url:
+        return transcribe_video(video_url)
 
 
 iface = gr.Interface(