US 12,230,247 B2
Systems and methods for improving content discovery in response to a voice query using a recognition rate which depends on detected trigger terms
Jeffry Copps Robert Jose, Tamil Nadu (IN); and Sindhuja Chonat Sri, Tamil Nadu (IN)
Assigned to Adeia Guides Inc., San Jose, CA (US)
Filed by Adeia Guides Inc., San Jose, CA (US)
Filed on Mar. 19, 2024, as Appl. No. 18/609,607.
Application 18/609,607 is a continuation of application No. 18/116,501, filed on Mar. 2, 2023, granted, now 11,961,507.
Application 18/116,501 is a continuation of application No. 16/889,036, filed on Jun. 1, 2020, granted, now 11,620,982, issued on Apr. 4, 2023.
Prior Publication US 2024/0221726 A1, Jul. 4, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G10L 15/02 (2006.01); G06F 16/432 (2019.01); G06F 16/438 (2019.01); G06F 40/279 (2020.01); G10L 15/14 (2006.01); G10L 15/26 (2006.01); H04M 3/51 (2006.01)
CPC G10L 15/02 (2013.01) [G06F 16/433 (2019.01); G06F 16/438 (2019.01); G06F 40/279 (2020.01); G10L 15/26 (2013.01); G10L 15/14 (2013.01); H04M 3/5116 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
generating a transcription of a query;
determining that the query was received within a threshold amount of time from a previous query, wherein the previous query comprises a term;
based on determining that the transcription includes a trigger term indicating that an intent of the query is to correct the previous query, temporarily increasing, for the term, a relaxation rate of an entity recognition model, wherein a number of interpretations considered for the term is based on the relaxation rate;
identifying a plurality of candidate entities to which the term refers using the entity recognition model and based on the increased relaxation rate;
performing a search based on the plurality of candidate entities; and
generating for output a search result of the search.