Update README.md

README.md

@@ -43,8 +43,8 @@ This dataset only has train split. To split out a test split, use
 ```python
 from datasets import load_dataset
 train_split = load_dataset("TingChen-ppmc/Shanghai_Dialect_Conversational_Speech_Corpus", split="train")
-# where test=0.5 denotes 0.5 of the dataset will be split to test split
-corpus = train_split.train_test_split(test=0.5)
+# where test_size=0.5 denotes 0.5 of the dataset will be split to test split
+corpus = train_split.train_test_split(test_size=0.5)
 ```
 
 A sample data would be