Multi-user authentication on a device

Patent metadata

Field	Value
Publication number	US-2020118550-A1
Application number	US-201916709132-A
Country	US
Kind code	A1
Filing date	Dec 10, 2019
Priority date	Apr 20, 2017
Publication date	Apr 16, 2020
Grant date	—

Abstract

In some implementations, authentication tokens corresponding to known users of a device are stored on the device. An utterance from a speaker is received. The utterance is classified as spoken by a particular known user of the known users. A query that includes a representation of the utterance and an indication of the particular known user as the speaker is provided using the authentication token of the particular known user.

First claim

What is claimed is:

1 . A computer-implemented method comprising:
receiving, by a device, an utterance from a speaker; classifying, by the device, the speaker of the utterance as not a known user of the device; providing, by the device and to a server, a query that includes authentication tokens that correspond to known users of the device, a representation of the utterance, and an indication that the speaker was classified as not a known user of the device; and receiving, by the device and from the server, a response to the query that is generated based on the authentication tokens that correspond to known users of the device, the representation of the utterance, and the indication that the speaker was classified as not a known user of the device.

2 . The method of claim 1 , wherein providing, by the device and to a server, a query that includes authentication tokens that correspond to known users of the device, a representation of the utterance, and an indication that the speaker was classified as not a known user of the device comprises:
generating the query to include, for each of the authentication tokens, a respective flag that indicates that the speaker was not classified as the known user that corresponds to the authentication token.

3 . The method of claim 1 , wherein providing, by the device and to a server, a query that includes authentication tokens that correspond to known users of the device, a representation of the utterance, and an indication that the speaker was classified as not a known user of the device comprises:
in response to classifying the speaker of the utterance as not a known user of the device, not setting flags that correspond to respective authentication tokens, where a flag set for an authentication token indicates that the speaker was classified as the known user that corresponds to the authentication token.

4 . The method of claim 1 , wherein providing, by the device and to a server, a query that includes authentication tokens that correspond to known users of the device, a representation of the utterance, and an indication that the speaker was classified as not a known user of the device comprises:
generating the query to include authentication tokens that correspond to all of the known users of the device.

5 . The method of claim 1 , wherein each of the authentication tokens indicate actions that may be performed in response to utterances.

6 . The method of claim 1 , wherein classifying, by the device, the speaker of the utterance as not a known user of the device comprises:
classifying, by the device, the speaker of the utterance as not a known user of the device based on speaker identification features of the utterance.

7 . The method of claim 1 , wherein the server is configured to perform operations of:
determining that the query does not indicate a known user as a speaker; in response to determining that the query does not indicate a known user as a speaker, determining whether the query is non-personal based on the representation of the utterance; in response to determining that the query is non-personal based on the representation of the utterance, determining the response to the query based on the authentication tokens included in the query; and providing the response to the device.

8 . A system comprising:
one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising: receiving, by a device, an utterance from a speaker; classifying, by the device, the speaker of the utterance as not a known user of the device; providing, by the device and to a server, a query that includes authentication tokens that correspond to known users of the device, a representation of the utterance, and an indication that the speaker was classified as not a known user of the device; and receiving, by the device and from the server, a response to the query that is generated based on the authentication tokens that correspond to known users of the device, the representation of the utterance, and the indication that the speaker was classified as not a known user of the device.

9 . The system of claim 8 , wherein providing, by the device and to a server, a query that includes authentication tokens that correspond to known users of the device, a representation of the utterance, and an indication that the speaker was classified as not a known user of the device comprises:
generating the query to include, for each of the authentication tokens, a respective flag that indicates that the speaker was not classified as the known user that corresponds to the authentication token.

10 . The system of claim 8 , wherein providing, by the device and to a server, a query that includes authentication tokens that correspond to known users of the device, a representation of the utterance, and an indication that the speaker was classified as not a known user of the device comprises:
in response to classifying the speaker of the utterance as not a known user of the device, not setting flags that correspond to respective authentication tokens, where a flag set for an authentication token indicates that the speaker was classified as the known user that corresponds to the authentication token.

11 . The system of claim 8 , wherein providing, by the device and to a server, a query that includes authentication tokens that correspond to known users of the device, a representation of the utterance, and an indication that the speaker was classified as not a known user of the device comprises:
generating the query to include authentication tokens that correspond to all of the known users of the device.

12 . The system of claim 8 , wherein each of the authentication tokens indicate actions that may be performed in response to utterances.

13 . The system of claim 8 , wherein classifying, by the device, the speaker of the utterance as not a known user of the device comprises:
classifying, by the device, the speaker of the utterance as not a known user of the device based on speaker identification features of the utterance.

14 . The system of claim 8 , wherein the server is configured to perform operations of:
determining that the query does not indicate a known user as a speaker; in response to determining that the query does not indicate a known user as a speaker, determining whether the query is non-personal based on the representation of the utterance; in response to determining that the query is non-personal based on the representation of the utterance, determining the response to the query based on the authentication tokens included in the query; and providing the response to the device.

15 . A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:
receiving, by a device, an utterance from a speaker; classifying, by the device, the speaker of the utterance as not a known user of the device; providing, by the device and to a server, a query that includes authentication tokens that correspond to known users of the device, a representation of the utterance, and an indication that the speaker was classified as not a known user of the device; and receiving, by the device and from the server, a response to the query that is generated based on the authentication tokens that correspond to known users of the device, the representation of the utterance, and the indication that the

Assignees

• Google Llc

Inventors

• Oktem Meltem
• Joglekar Taral Pradeep
• Heryandi Fnu
• Chao Pu-Sen
• Lopez Moreno Ignacio
• Rajadhyaksha Salil
• Gruenstein Alexander H
• Melendo Casado Diego

Classifications

• G10L15/08 Primary

  Speech classification or search · CPC title

• G10L15/22

  Procedures used during a speech recognition process, e.g. man-machine dialogue · CPC title

• G06F16/636

  by using biological or physiological data · CPC title

• G10L2015/088

  Word spotting · CPC title

• G10L17/06 Primary

  Decision making techniques; Pattern matching strategies · CPC title