diff --git a/test/test_datasets.py b/test/test_datasets.py
index 4d8cacbb4a..947dfa5fd6 100644
--- a/test/test_datasets.py
+++ b/test/test_datasets.py
@@ -1,11 +1,7 @@
 import os
 import unittest
 
-from torchaudio.datasets.commonvoice import COMMONVOICE
-from torchaudio.datasets.librispeech import LIBRISPEECH
-from torchaudio.datasets.utils import DiskCache
-from torchaudio.datasets.vctk import VCTK
-from torchaudio.datasets.yesno import YESNO
+from torchaudio.datasets import COMMONVOICE, LIBRISPEECH, VCTK, YESNO, DiskCache
 
 
 class TestDatasets(unittest.TestCase):
diff --git a/torchaudio/datasets/__init__.py b/torchaudio/datasets/__init__.py
index ef395a0ca4..848ed7202a 100644
--- a/torchaudio/datasets/__init__.py
+++ b/torchaudio/datasets/__init__.py
@@ -2,6 +2,6 @@
 from .librispeech import LIBRISPEECH
 from .vctk import VCTK
 from .yesno import YESNO
-from .utils import DiskCache
+from .cache import DiskCache
 
 __all__ = ("COMMONVOICE", "LIBRISPEECH", "VCTK", "YESNO", "DiskCache")
diff --git a/torchaudio/datasets/cache.py b/torchaudio/datasets/cache.py
new file mode 100644
index 0000000000..71cb150825
--- /dev/null
+++ b/torchaudio/datasets/cache.py
@@ -0,0 +1,36 @@
+import os
+
+import torch
+from torch.utils.data import Dataset
+from torchaudio.datasets.utils import makedir_exist_ok
+
+
+class DiskCache(Dataset):
+    """
+    Wrap a dataset so that, whenever a new item is returned, it is saved to disk.
+    """
+
+    def __init__(self, dataset, location=".cached"):
+        self.dataset = dataset
+        self.location = location
+
+        self._id = id(self)
+        self._cache = [None] * len(dataset)
+
+    def __getitem__(self, n):
+        if self._cache[n]:
+            f = self._cache[n]
+            return torch.load(f)
+
+        f = str(self._id) + "-" + str(n)
+        f = os.path.join(self.location, f)
+        item = self.dataset[n]
+
+        self._cache[n] = f
+        makedir_exist_ok(self.location)
+        torch.save(item, f)
+
+        return item
+
+    def __len__(self):
+        return len(self.dataset)
diff --git a/torchaudio/datasets/utils.py b/torchaudio/datasets/utils.py
index 25a30293aa..a57b6c1c1f 100644
--- a/torchaudio/datasets/utils.py
+++ b/torchaudio/datasets/utils.py
@@ -11,7 +11,6 @@
 import six
 import torch
 import torchaudio
-from torch.utils.data import Dataset
 from torch.utils.model_zoo import tqdm
 
 
@@ -190,34 +189,3 @@ def walk_files(root, suffix, prefix=False, remove_suffix=False):
                     f = os.path.join(root, f)
 
                 yield f
-
-
-class DiskCache(Dataset):
-    """
-    Wrap a dataset so that, whenever a new item is returned, it is saved to disk.
-    """
-
-    def __init__(self, dataset, location=".cached"):
-        self.dataset = dataset
-        self.location = location
-
-        self._id = id(self)
-        self._cache = [None] * len(dataset)
-
-    def __getitem__(self, n):
-        if self._cache[n]:
-            f = self._cache[n]
-            return torch.load(f)
-
-        f = str(self._id) + "-" + str(n)
-        f = os.path.join(self.location, f)
-        item = self.dataset[n]
-
-        self._cache[n] = f
-        makedir_exist_ok(self.location)
-        torch.save(item, f)
-
-        return item
-
-    def __len__(self):
-        return len(self.dataset)