debug

src/se_extractor.py

@@ -13,6 +13,7 @@ model_size = 'medium'
 def split_audio_whisper(audio_path, target_dir='processed',needs_offset=True):
     print("in whisper split")
     model = WhisperModel('medium', device="cuda", compute_type="float16")
+    print("loaded")
     audio = AudioSegment.from_file(audio_path)
     max_len = len(audio)
 
@@ -21,7 +22,7 @@ def split_audio_whisper(audio_path, target_dir='processed',needs_offset=True):
     
     segments, info = model.transcribe(audio_path, beam_size=5, word_timestamps=True)
     segments = list(segments)    
-
+    print(segments)
     # create directory
     os.makedirs(target_folder, exist_ok=True)
     wavs_folder = os.path.join(target_folder, 'wavs')
@@ -134,7 +135,7 @@ def get_se(audio_path, vc_model, target_dir='processed', vad=True):
         wavs_folder = split_audio_vad(audio_path, target_dir)
     else:
         wavs_folder = split_audio_whisper(audio_path, target_dir)
-    
+    print("Done")
     audio_segs = glob(f'{wavs_folder}/*.wav')
     if len(audio_segs) == 0:
         raise NotImplementedError('No audio segments found!')