US 12,437,161 B2
Providing a well-formed alternate phrase as a suggestion in lieu of a not well-formed phrase
Wangqing Yuan, Wilmington, MA (US); David Kogan, Natick, MA (US); Vincent Lacey, Mountain View, CA (US); Guanglei Wang, Cambridge, CA (US); Shaun Post, San Mateo, CA (US); Bryan Christopher Horling, Belmont, MA (US); and Michael Anthony Schuler, San Francisco, CA (US)
Assigned to GOOGLE LLC, Mountain View, CA (US)
Filed by GOOGLE LLC, Mountain View, CA (US)
Filed on May 23, 2024, as Appl. No. 18/673,145.
Application 18/673,145 is a continuation of application No. 17/351,872, filed on Jun. 18, 2021, granted, now 12,019,999.
Prior Publication US 2024/0311585 A1, Sep. 19, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 40/51 (2020.01); G06F 40/211 (2020.01); G06F 40/289 (2020.01)
CPC G06F 40/51 (2020.01) [G06F 40/211 (2020.01); G06F 40/289 (2020.01)] 15 Claims
OG exemplary drawing
 
1. A system comprising:
at least one processor; and
memory storing instructions that, when executed, cause the least one processor to be operable to:
prior to receiving audio data that captures a spoken query, of a user, that includes a particular phrase in a particular language:
determine that the particular phrase that is in the particular language is not well formed;
in response to determining that the particular phrase in the particular language is not well formed:
process the particular phrase in the particular language to identify a plurality of alternate phrases that are each in the particular language, that are each indicated as well-formed, and that each have a defined relationship to the particular phrase;
generate a corresponding score for each of the alternate phrases; and
select, based on the corresponding scores for each of the alternate phrases, a particular alternate phrase of the alternate phrases, that is in the particular language that is well formed;
subsequent to selecting the particular alternate phrase:
receive a representation of the spoken query, of the user, that includes the particular phrase, wherein the spoken query is provided by the user via an automated assistant interface of a client device of the user, and wherein the spoken query is not in a primary language for the user;
in response to determining that the representation of the spoken query includes the particular phrase, in response to selecting the particular alternate phrase, and in response to determining that the particular language, of the particular phrase, is not specified as the primary language for the user:
refrain from causing a particular language response, in the particular language, to be visually rendered at the client device;
cause the particular alternate phrase to be visually rendered, in the particular language, at the client device as a suggestion to be utilized in lieu of the particular phrase; and
in response to receiving a user selection of the particular alternate phrase:
cause the particular language response, that is responsive to the particular alternate phrase in the particular language, to be visually rendered at the client device; and
cause a primary language response, that is responsive to the particular alternate phrase in the primary language, to be visually rendered at the client device and along with the particular language response.