Patent US 11,076,091

Features are disclosed for interacting with a user to take a photo when the user is actually ready. The features combine computer vision and voice recognition to determine when the user is ready. Additional features to interact with the user to compose the image are also described. For example, the system may play audio requesting a response (e.g., “move a bit to the left” or “say cheese”). The response may include a user utterance or physical movement detected by the system. Based on the response, the image may be taken.