Update create_dataset.py

create_dataset.py

@@ -90,14 +90,14 @@ def extract_vocal_demucs(model, filename, out_filename, sr=44100, device=None, s
 
 
 def main(
-    clips_csv_filepath = "theovon.csv",
-    character = "theovon",
+    clips_csv_filepath = "data.csv",
+    character = "somebody",
     do_extract_vocals = False,
     whisper_size = "medium",
     # Where raw yt clips will be downloaded to
-    dl_dir = "raw_data",
+    dl_dir = "downloads",
     # Where actual data will be organized
-    data_dir = "prepared_data",
+    data_dir = "dataset_raw",
 ):
     dl_path = Path(dl_dir) / character
     dl_path.mkdir(exist_ok=True, parents=True)