diff --git a/machine/translation/huggingface/hugging_face_nmt_model_trainer.py b/machine/translation/huggingface/hugging_face_nmt_model_trainer.py
index 1192243e..2d8641c7 100644
--- a/machine/translation/huggingface/hugging_face_nmt_model_trainer.py
+++ b/machine/translation/huggingface/hugging_face_nmt_model_trainer.py
@@ -154,6 +154,9 @@ def train(
         tgt_lang = self._tgt_lang
         if tgt_lang is None:
             tgt_lang = "tgt"
+        if src_lang == tgt_lang:
+            src_lang += "_src"
+            tgt_lang += "_trg"
 
         if isinstance(self._corpus, Dataset):
             train_dataset = self._corpus
@@ -206,13 +209,12 @@ def add_tokens(tokenizer: Any, missing_tokens: List[str]) -> Any:
                     use_fast=True,
                 )
                 # using unofficially supported behavior to set the normalizer
+                lang_codes = []
                 tokenizer.backend_tokenizer.normalizer = norm_tok.backend_tokenizer.normalizer  # type: ignore
-                if self._add_unk_src_tokens and self._add_unk_tgt_tokens:
-                    lang_codes = [src_lang, tgt_lang]
-                elif self._add_unk_src_tokens:
-                    lang_codes = [src_lang]
-                else:
-                    lang_codes = [tgt_lang]
+                if self._add_unk_src_tokens:
+                    lang_codes.append(src_lang)
+                if self._add_unk_tgt_tokens:
+                    lang_codes.append(tgt_lang)
                 missing_tokens = find_missing_characters(tokenizer, train_dataset, lang_codes)
                 if missing_tokens:
                     tokenizer = add_tokens(tokenizer, missing_tokens)