modified sort transcript function to support a new format input

utlis.py

@@ -22,10 +22,16 @@ def sort_transcript(file_path, save_path):
     file_ext = os.path.splitext(file_path)[1]
     
     if file_ext == '.txt' or file_ext == '.tsv':
-        # read the tab-separated plaintext file into a DataFrame
-        table = pd.read_csv(file_path, sep='\t', header=None, \
-                            names=['Speaker', 'Empty', 'Start', 'Start (s)', 'End', 'End (s)', 'Duration', 'Duration (s)', 'Transcript'], 
-                            index_col=False)
+        with open(file_path, 'r') as file:
+            # Read the lines from the file, ignoring lines starting with '#'
+            lines = [line.strip() for line in file if line.strip() and not line.startswith('''"#file''')]
+        # Create a DataFrame from the remaining lines
+        table = pd.DataFrame([line.split('\t') for line in lines])
+        # since there two formats of input files
+        if len(table.columns) == 5:
+            table.columns = ['Speaker', 'Empty', 'Start', 'End', 'Transcript']
+        else:
+            table.columns = ['Speaker', 'Empty', 'Start', 'Start (s)', 'End', 'End (s)', 'Duration', 'Duration (s)', 'Transcript']
     else:
         raise ValueError("Unsupported file format. Must be '.txt' or '.tsv'.")