diff --git a/paddlemix/appflow/text2speech_synthesize.py b/paddlemix/appflow/text2speech_synthesize.py
index 5d33e23b82878..8d348df0e005d 100644
--- a/paddlemix/appflow/text2speech_synthesize.py
+++ b/paddlemix/appflow/text2speech_synthesize.py
@@ -16,6 +16,26 @@
 from .apptask import AppTask
 import paddle
 from paddlemix.utils.log import logger
+import paddle.nn as nn, paddle
+def get_parameter_dtype(parameter: nn.Layer) -> paddle.dtype:
+    try:
+        return next(parameter.named_parameters())[1].dtype
+    except StopIteration:
+        try:
+            return next(parameter.named_buffers())[1].dtype
+        except StopIteration:
+            return parameter._dtype
+@property
+def dtype_getter(self):
+    if hasattr(self, "__dtype"):
+        return self.__dtype
+    return get_parameter_dtype(self)
+nn.Layer.dtype = dtype_getter
+
+@nn.Layer.dtype.setter
+def dtype_setter(self, value):
+    self.__dtype = value
+nn.Layer.dtype = dtype_setter
 
 
 class AudioTTSTask(AppTask):
@@ -24,6 +44,7 @@ def __init__(self, task, model, **kwargs):
         
         # Default to static mode
         self._static_mode = False
+        
         self._construct_model()
 
     def _construct_model(self):