diff --git a/egs/vctk/TTS/vits/train.py b/egs/vctk/TTS/vits/train.py
index e306a9670..3fc637a46 100755
--- a/egs/vctk/TTS/vits/train.py
+++ b/egs/vctk/TTS/vits/train.py
@@ -452,7 +452,7 @@ def train_one_epoch(
                     feats_lengths=features_lens,
                     speech=audio,
                     speech_lengths=audio_lens,
-                    sid=speakers,
+                    sids=speakers,
                     forward_generator=False,
                 )
             for k, v in stats_d.items():
@@ -471,7 +471,7 @@ def train_one_epoch(
                     feats_lengths=features_lens,
                     speech=audio,
                     speech_lengths=audio_lens,
-                    sid=speakers,
+                    sids=speakers,
                     forward_generator=True,
                     return_sample=params.batch_idx_train % params.log_interval == 0,
                 )
@@ -652,7 +652,7 @@ def compute_validation_loss(
                 feats_lengths=features_lens,
                 speech=audio,
                 speech_lengths=audio_lens,
-                sid=speakers,
+                sids=speakers,
                 forward_generator=False,
             )
             assert loss_d.requires_grad is False
@@ -667,7 +667,7 @@ def compute_validation_loss(
                 feats_lengths=features_lens,
                 speech=audio,
                 speech_lengths=audio_lens,
-                sid=speakers,
+                sids=speakers,
                 forward_generator=True,
             )
             assert loss_g.requires_grad is False
@@ -742,7 +742,7 @@ def scan_pessimistic_batches_for_oom(
                     feats_lengths=features_lens,
                     speech=audio,
                     speech_lengths=audio_lens,
-                    sid=speakers,
+                    sids=speakers,
                     forward_generator=False,
                 )
             optimizer_d.zero_grad()
@@ -756,7 +756,7 @@ def scan_pessimistic_batches_for_oom(
                     feats_lengths=features_lens,
                     speech=audio,
                     speech_lengths=audio_lens,
-                    sid=speakers,
+                    sids=speakers,
                     forward_generator=True,
                 )
             optimizer_g.zero_grad()