diff --git a/tests/transformers_tests/models/seamless_m4t/test_modeling_seamless_m4t.py b/tests/transformers_tests/models/seamless_m4t/test_modeling_seamless_m4t.py
index b31ee36d0a..cdf915ee1d 100644
--- a/tests/transformers_tests/models/seamless_m4t/test_modeling_seamless_m4t.py
+++ b/tests/transformers_tests/models/seamless_m4t/test_modeling_seamless_m4t.py
@@ -230,6 +230,7 @@ def get_config(self):
             inputs_dict_speech["labels"],
         ),
         {},
+        [1, 2, 3],
         {
             "logits": 0,
             "encoder_last_hidden_state": 2,
@@ -248,6 +249,7 @@ def get_config(self):
             inputs_dict_speech["labels"],
         ),
         {},
+        [1, 2, 3],
         {
             "logits": 0,
             "encoder_last_hidden_state": 2,
@@ -266,6 +268,7 @@ def get_config(self):
             inputs_dict_text["labels"],
         ),
         {},
+        [0, 1, 2, 3],
         {
             "logits": 0,
             "encoder_last_hidden_state": 2,
@@ -284,6 +287,7 @@ def get_config(self):
             inputs_dict_text["labels"],
         ),
         {},
+        [0, 1, 2, 3],
         {
             "logits": 0,
             "encoder_last_hidden_state": 2,
@@ -303,6 +307,7 @@ def get_config(self):
             inputs_dict_text["labels"],
         ),
         {},
+        [0, 2, 3, 4],
         {
             "logits": 0,
             "encoder_last_hidden_state": 2,
@@ -312,7 +317,7 @@ def get_config(self):
 
 
 @pytest.mark.parametrize(
-    "name,pt_module,ms_module,init_args,init_kwargs,inputs_args,inputs_kwargs,outputs_map,dtype,mode",
+    "name,pt_module,ms_module,init_args,init_kwargs,inputs_args,inputs_kwargs,inputs_type_idx,outputs_map,dtype,mode",
     [
         case
         + [
@@ -334,6 +339,7 @@ def test_named_modules(
     init_kwargs,
     inputs_args,
     inputs_kwargs,
+    inputs_type_idx,
     outputs_map,
     dtype,
     mode,
@@ -357,6 +363,16 @@ def test_named_modules(
         ms_inputs_kwargs.update({"hidden_dtype": MS_DTYPE_MAPPING[ms_dtype]})
     ms_inputs_kwargs["return_dict"] = False
 
+    pt_inputs_args = tuple(
+        tensor.long() if i in inputs_type_idx else tensor.to(PT_DTYPE_MAPPING[pt_dtype])
+        for i, tensor in enumerate(pt_inputs_args)
+    )
+
+    ms_inputs_args = tuple(
+        tensor.to(ms.int64) if i in inputs_type_idx else tensor.to(MS_DTYPE_MAPPING[ms_dtype])
+        for i, tensor in enumerate(ms_inputs_args)
+    )
+
     with torch.no_grad():
         pt_outputs = pt_model(*pt_inputs_args, **pt_inputs_kwargs)
     ms_outputs = ms_model(*ms_inputs_args, **ms_inputs_kwargs)