diff --git a/tensorflow/lite/schema/schema_v0.fbs b/tensorflow/lite/schema/schema_v0.fbs
index a080bbdaab4..e543df613cb 100644
--- a/tensorflow/lite/schema/schema_v0.fbs
+++ b/tensorflow/lite/schema/schema_v0.fbs
@@ -35,7 +35,7 @@ table QuantizationParameters {
 
 table Tensor {
   // The tensor shape. The meaning of each entry is operator-specific but
-  // builtin ops use: [batch size, number of channels, height, width] (That's
+  // builtin ops use: [batch size, height, width, number of channels] (That's
   // Tensorflow's NHWC).
   shape:[int];
   type:TensorType;
diff --git a/tensorflow/lite/schema/schema_v1.fbs b/tensorflow/lite/schema/schema_v1.fbs
index 779492e8ee3..d49ea8e7f05 100644
--- a/tensorflow/lite/schema/schema_v1.fbs
+++ b/tensorflow/lite/schema/schema_v1.fbs
@@ -40,7 +40,7 @@ table QuantizationParameters {
 
 table Tensor {
   // The tensor shape. The meaning of each entry is operator-specific but
-  // builtin ops use: [batch size, number of channels, height, width] (That's
+  // builtin ops use: [batch size, height, width, number of channels] (That's
   // Tensorflow's NHWC).
   shape:[int];
   type:TensorType;
diff --git a/tensorflow/lite/schema/schema_v2.fbs b/tensorflow/lite/schema/schema_v2.fbs
index 94963a4f9c2..05464a7ea60 100644
--- a/tensorflow/lite/schema/schema_v2.fbs
+++ b/tensorflow/lite/schema/schema_v2.fbs
@@ -41,7 +41,7 @@ table QuantizationParameters {
 
 table Tensor {
   // The tensor shape. The meaning of each entry is operator-specific but
-  // builtin ops use: [batch size, number of channels, height, width] (That's
+  // builtin ops use: [batch size, height, width, number of channels] (That's
   // Tensorflow's NHWC).
   shape:[int];
   type:TensorType;
diff --git a/tensorflow/lite/schema/schema_v3.fbs b/tensorflow/lite/schema/schema_v3.fbs
index 3b3c763ffbc..86d561a408a 100644
--- a/tensorflow/lite/schema/schema_v3.fbs
+++ b/tensorflow/lite/schema/schema_v3.fbs
@@ -47,7 +47,7 @@ table QuantizationParameters {
 
 table Tensor {
   // The tensor shape. The meaning of each entry is operator-specific but
-  // builtin ops use: [batch size, number of channels, height, width] (That's
+  // builtin ops use: [batch size, height, width, number of channels] (That's
   // Tensorflow's NHWC).
   shape:[int];
   type:TensorType;