US 12,451,133 B2
Voice-based menu personalization
Jodessiah Sumpter, Alpharetta, GA (US); Christian McDaniel, Atlanta, GA (US); Kendall Marie Rose, Canton, MI (US); and Shaundell D. Thompson, Loganville, GA (US)
Assigned to NCR Voyix Corporation, Atlanta, GA (US)
Filed by NCR Voyix Corporation, Atlanta, GA (US)
Filed on Mar. 23, 2023, as Appl. No. 18/125,443.
Application 18/125,443 is a continuation of application No. 17/105,159, filed on Nov. 25, 2020, granted, now 11,676,592.
Prior Publication US 2023/0230590 A1, Jul. 20, 2023
Int. Cl. G10L 15/22 (2006.01); G10L 15/26 (2006.01); G06Q 30/0601 (2023.01); G06Q 50/12 (2012.01)
CPC G10L 15/22 (2013.01) [G10L 15/26 (2013.01); G06Q 30/0635 (2013.01); G06Q 50/12 (2013.01); G10L 2015/227 (2013.01)] 10 Claims
OG exemplary drawing
 
1. A method, comprising:
engaging a consumer in a natural language voice dialogue to place an order at an establishment using voice characteristics specific to the consumer and a lexicon specific to ordering with the establishment;
translating speech of the consumer into text based on the voice characteristics that are specific to the consumer and the lexicon associated with ordering, wherein the translating utilizes a machine learning algorithm that receives as input the voice characteristics, an audio snippet of consumer speech, and the lexicon, and provides as output the text that corresponds to the speech;
customizing voice interaction using deep learning techniques by performing an initial voice training session to capture voice features and characteristics for the consumer;
wherein during the initial voice training session:
displaying text sentences on a display device;
receiving audio snippets of the speech from the consumer in response to prompts to the consumer to repeat the text sentences; and
training the machine learning algorithm to return voice features of the consumer, wherein corresponding features are retained in a voice profile specific to the consumer; and
training a non-user specific machine learning algorithm to utilize the voice profile to enable text translation for the consumer;
placing the order within an order system associated with the establishment using menu options obtained from the consumer during the engaging; and
sending a message to a fulfillment terminal of the establishment to begin order preparation when location data associated with a consumer device of the consumer indicates that the consumer is in route to the establishment to pick up the order and a calculated time in which the consumer is estimated to arrive at the establishment, wherein the calculated time corresponds with an order preparation time for the establishment to complete the order.