add decoding to get correct swedish chars

run_speech_recognition_ctc.py

@@ -521,9 +521,9 @@ def main():
 
     def remove_special_characters(batch):
         if chars_to_ignore_regex is not None:
-            batch["target_text"] = re.sub(chars_to_ignore_regex, "", batch[text_column_name]).replace("\\\\Punkt").replace("\\\\Komma").lower() + " "
+            batch["target_text"] = re.sub(chars_to_ignore_regex, "", batch[text_column_name]).replace("\\\\Punkt").replace("\\\\Komma").lower().decode("utf-8") + " "
         else:
-            batch["target_text"] = batch[text_column_name].replace("\\\\Punkt").replace("\\\\Komma").lower() + " "
+            batch["target_text"] = batch[text_column_name].replace("\\\\Punkt").replace("\\\\Komma").lower().decode("utf-8") + " "
         return batch
 
     num_workers = data_args.preprocessing_num_workers