US 12,169,681 B2
Context-aware font recommendation from text
Amirreza Shirani, Cupertino, CA (US); Franck Dernoncourt, San Jose, CA (US); Jose Ignacio Echevarria Vallespi, Belmont, CA (US); Paul Asente, Redwood City, CA (US); Nedim Lipka, Campbell, CA (US); and Thamar I. Solorio Martinez, Cypress, TX (US)
Assigned to Adobe Inc., San Jose, CA (US)
Filed by Adobe Inc., San Jose, CA (US)
Filed on Sep. 29, 2021, as Appl. No. 17/489,474.
Claims priority of provisional application 63/184,182, filed on May 4, 2021.
Prior Publication US 2022/0358280 A1, Nov. 10, 2022
Int. Cl. G06F 40/109 (2020.01); G06F 40/169 (2020.01); G06N 3/02 (2006.01)
CPC G06F 40/109 (2020.01) [G06F 40/169 (2020.01); G06N 3/02 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
receiving a selection of text;
providing the selection of text to a font recommendation model trained to predict a font recommendation by learning associations between visual attributes of a plurality of font types and verbal contexts of training texts;
generating, by the font recommendation model, an embedding representing a verbal context of the selection of text, the verbal context of the selection of text including emotions conveyed by the selection of text;
generating, by the font recommendation model, a prediction score for each of the plurality of font types based on the embedding representing the verbal context of the selection of text, wherein each prediction score is based on a relationship between visual attributes of a corresponding font type and the emotions conveyed by the selection of text, wherein a higher prediction score for a font type indicates greater congruency between the font type and the emotions conveyed by the selection of text; and
returning at least one recommended font type based on the prediction score for each of the plurality of font types.