diff --git a/TTS/tts/models/vits.py b/TTS/tts/models/vits.py
index f91115be84..f3b8074023 100644
--- a/TTS/tts/models/vits.py
+++ b/TTS/tts/models/vits.py
@@ -1472,7 +1472,6 @@ def format_batch(self, batch: Dict) -> Dict:
 
         if speaker_ids is not None:
             speaker_ids = torch.LongTensor(speaker_ids)
-            batch["speaker_ids"] = speaker_ids
 
         # get d_vectors from audio file names
         if self.speaker_manager is not None and self.speaker_manager.embeddings and self.args.use_d_vector_file: