Method and system for selecting a voice assistant

Patent metadata

Field	Value
Publication number	US-12444422-B2
Application number	US-202218090081-A
Country	US
Kind code	B2
Filing date	Dec 28, 2022
Priority date	Dec 28, 2022
Publication date	Oct 14, 2025
Grant date	Oct 14, 2025

Abstract

A method for processing voice input is disclosed. The method may be performed by a device including a voice assistant manager and a plurality of voice assistants. In some embodiments, the method includes receiving an utterance from a user, detecting a category of the utterance, and communicating the utterance to a selected voice assistant of the plurality of voice assistants. The selected voice assistant may be associated with the detected category. In some embodiments, the selected voice assistant may generate a response to utterance, and the response may be output to the user.

First claim

What is claimed is:

1. A method for processing voice input from a user, the method comprising:
receiving an utterance from the user at a computing device;
detecting a wake word in the utterance;
based at least in part on the wake word, identifying a called assistant from a plurality of voice assistants;
communicating the utterance to the called assistant;
determining a category of the utterance from a plurality of categories, wherein determining the category of the utterance is performed in response to receiving an error from the called assistant;
selecting an assistant from the plurality of voice assistants;
communicating the utterance to the selected assistant; and
transmitting a response from the selected assistant to the user,
wherein selecting the assistant from the plurality of voice assistants comprises determining that the selected assistant is associated with the category.

2. The method of claim 1 , further comprising detecting a subcategory of the utterance;
wherein selecting the assistant from the plurality of voice assistants further comprises determining that the selected assistant is associated with the subcategory.

3. The method of claim 1 , further comprising detecting an action of the utterance;
wherein selecting the assistant from the plurality of voice assistants further comprises determining that the selected assistant is associated with the action.

4. The method of claim 1 , further comprising, prior to communicating the utterance to the selected assistant:
transmitting a communication to the user, wherein the communication requests permission to transmit the utterance to the selected assistant; and
receiving a confirmation from the user.

5. The method of claim 1 , further comprising:
identifying, from the plurality of voice assistants, a plurality of voice assistants associated with the category; and
communicating the utterance to each of the plurality of voice assistants associated with the category.

6. The method of claim 1 , wherein selecting the assistant from the plurality of voice assistants further comprises:
identifying, from the plurality of voice assistants, multiple voice assistants associated with the category; and
selecting the assistant from the multiple voice assistants associated with the category.

7. The method of claim 6 , wherein selecting the assistant from the multiple voice assistants associated with the category comprises selecting the assistant based on a popularity of the selected assistant at a time of day.

8. The method of claim 1 , further comprising receiving a category customization input;
wherein the category customization input includes:
a customized category; and
one or more voice assistants of the plurality of voice assistants.

9. The method of claim 1 , further comprising, prior to selecting the assistant from the plurality of voice assistants, receiving a subscription request from the selected assistant, the subscription request including one or more categories associated with the selected assistant.

10. The method of claim 1 , wherein each voice assistant of the plurality of voice assistants is installed on the computing device.

11. The method of claim 1 , wherein determining the category of the utterance comprises inputting the utterance into a category detection model; and
wherein the category detection model is a machine learning model trained to recognize one or more categories of the plurality of categories.

12. A device for processing voice input, the device comprising:
a processor; and
memory storing instructions that, when executed by the processor, cause the device to:
receive an utterance from a user;
detect a wake word in the utterance;
based at least in part on the wake word, identify a called assistant from a plurality of voice assistants;
communicate the utterance to the called assistant;
determine, from a plurality of categories, a category of the utterance, wherein determining the category of the utterance is performed in response to receiving an error from the called assistant;
select an assistant from the plurality of voice assistants; and
communicate the utterance to the selected assistant,
wherein selecting the assistant from the plurality of assistants comprises determining that the selected assistant is associated with the category.

13. The device of claim 12 , wherein the instructions, when executed by the processor, further cause the device to, prior to communicating the utterance to the selected assistant, transmit a communication to the user, the communication including information related to the selected assistant.

14. A device for processing a voice utterance, the device comprising:
a voice assistant manager;
a plurality of voice assistants;
a processor; and
memory coupled to the processor, wherein the memory stores instructions that, when executed by the processor, cause the voice assistant manager to:
receive the utterance from a user;
detect a wake word in the utterance;
based at least in part on the wake word, identify a called assistant from a plurality of voice assistants;
communicate the utterance to the called assistant;
determine a category of the utterance, wherein determining the category of the utterance is performed in response to receiving an error from the called assistant;
select an assistant from the plurality of voice assistants, wherein the assistant is associated with the category; and
communicate the utterance to the selected assistant,

wherein the instructions, when executed by the processor, cause the selected assistant to receive the utterance and to generate a response to the utterance.

15. The device of claim 14 , wherein the instructions, when executed by the processor, further cause the voice assistant manager to select, from the plurality of voice assistants, multiple voice assistants associated with the category; and
wherein selecting the assistant from the plurality of voice assistants comprises selecting the assistant from the multiple voice assistants associated with the category.

Assignees

• Spotify Ab

Inventors

• Bromand Daniel
• Roth Björn Erik

Classifications

• G10L15/22 Primary

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

• G10L15/01

  Assessment or evaluation of speech recognition systems · CPC title

• G10L2015/088

  Word spotting · CPC title

• G10L15/08

  Speech classification or search · CPC title

• H04L51/214

  using selective forwarding · CPC title