diff --git a/mediapipe/tasks/cc/vision/image_segmenter/image_segmenter.h b/mediapipe/tasks/cc/vision/image_segmenter/image_segmenter.h
index 43bf5b7e6..511d3b9c1 100644
--- a/mediapipe/tasks/cc/vision/image_segmenter/image_segmenter.h
+++ b/mediapipe/tasks/cc/vision/image_segmenter/image_segmenter.h
@@ -98,7 +98,7 @@ struct ImageSegmenterOptions {
 //   - list of segmented masks.
 //   - if `output_type` is CATEGORY_MASK, uint8 Image, Image vector of size 1.
 //   - if `output_type` is CONFIDENCE_MASK, float32 Image list of size
-//     `cahnnels`.
+//     `channels`.
 //   - batch is always 1
 // An example of such model can be found at:
 // https://tfhub.dev/tensorflow/lite-model/deeplabv3/1/metadata/2
diff --git a/mediapipe/tasks/python/vision/image_segmenter.py b/mediapipe/tasks/python/vision/image_segmenter.py
index 9ef911f75..62fc8bb7c 100644
--- a/mediapipe/tasks/python/vision/image_segmenter.py
+++ b/mediapipe/tasks/python/vision/image_segmenter.py
@@ -110,7 +110,7 @@ class ImageSegmenter(base_vision_task_api.BaseVisionTaskApi):
     - list of segmented masks.
     - if `output_type` is CATEGORY_MASK, uint8 Image, Image vector of size 1.
     - if `output_type` is CONFIDENCE_MASK, float32 Image list of size
-      `cahnnels`.
+      `channels`.
     - batch is always 1
 
   An example of such model can be found at: