Update app.py

app.py

@@ -52,13 +52,13 @@ def process_video(video_file, language_choice):
     audio = AudioSegment.from_file(reference_audio)
     resampled_audio = audio.set_frame_rate(48000)
     resampled_audio.export(reference_audio, format="wav")
-    audio_clip = AudioFileClip(reference_audio)
 
     # Enhance the audio
     model, df_state, _ = init_df()
     audio, _ = load_audio(reference_audio, sr=df_state.sr())
     enhanced = enhance(model, df_state, audio)
     save_audio(reference_audio, enhanced, df_state.sr())
+    audio_clip = AudioFileClip(reference_audio)
 
     src_path = os.path.join(output_dir, "tmp.wav")
 
@@ -158,7 +158,7 @@ def process_video(video_file, language_choice):
                 start, end, translated_text = segment
                 segment_path = os.path.join(output_dir, f'segment_{start}_{end}.wav')
                 model.tts_to_file(translated_text, speaker_id, segment_path, speed=speed)
-                
+                print(start)
                 reference_speaker = AudioFileClip.subclip(audio_clip, int(start), int(end))  # This is the voice you want to clone
                 reference_speaker.write_audiofile("reference_speaker.wav")
                 target_se, audio_name = se_extractor.get_se("reference_speaker.wav", tone_color_converter, vad=False)