Spoken Utterance Fulfillment Using User-Specific Action Selection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Automated assistants often implement incorrect fulfillment actions for spoken utterances, leading to wasted computational and natural resources due to mismatched user preferences, as they lack the ability to dynamically adapt responses based on individual user settings and contextual scenarios.

Innovation Solution

A system that identifies users and adapts fulfillment actions by using user-specific models and rules, processing spoken utterances through ASR and NLU, and selecting appropriate actions based on user identification and contextual signals to provide personalized responses.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If automated assistants implement standard fulfillment actions for spoken utterances, then the system can respond to user commands, but the fulfillment actions may not match individual user preferences leading to incorrect responses

Engineering Contradiction:

Improveaccuracy of fulfillment actionsVSAvoidcomplexity of user identification and adaptation system

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system segments the fulfillment process by creating separate fulfillment models for different users. Each user has their own customized fulfillment model that stores user-specific preferences and behaviors. When a user provides a spoken utterance, the system identifies the user and retrieves their specific fulfillment model, thereby segmenting the general fulfillment process into user-specific instances that accurately reflect individual preferences.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary actions by pre-collecting user interaction data and pre-training customized fulfillment models for each user before actual fulfillment is needed. User profiles and preferences are established in advance through initial interactions, so when a spoken utterance is received, the system can immediately apply the pre-prepared user-specific model without requiring real-time analysis of user preferences.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If automated assistants use user-specific fulfillment models, then fulfillment accuracy improves, but computational resources are wasted on incorrect fulfillment actions and additional interactions

Engineering Contradiction:

Improveaccuracy of fulfillment actionsVSAvoidcomputational resource wastage

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The system performs preliminary action by pre-training user-specific fulfillment models during off-peak times based on collected user interaction data. This allows the models to be ready in advance, reducing the need for computationally intensive real-time analysis when processing spoken utterances. The preliminary preparation of user profiles and preferences minimizes runtime computational overhead.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements feedback mechanisms where user responses to fulfillment actions are continuously monitored and used to refine and update user-specific fulfillment models. This feedback loop allows the system to learn from incorrect or successful fulfillments, progressively improving accuracy and reducing the frequency of incorrect actions that waste computational resources.

Inventive Principle:

Principle #23Feedback

3Ease of operation

If automated assistants implement generic fulfillment actions, then the system operation is simple, but user preferences are not respected requiring manual correction

Engineering Contradiction:

Improveease of system operationVSAvoidtime for manual correction and additional interactions

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The system segments fulfillment operations into user-specific instances by maintaining separate fulfillment models for different users. Each model encapsulates that user's preferences, making the system operation simple from the user's perspective while internally handling the complexity of personalized fulfillment. Users experience streamlined interaction without manual corrections because their specific preferences are automatically applied.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system enables self-service by automatically adapting fulfillment actions based on user identification and stored user profiles. The system serves itself by autonomously selecting appropriate fulfillment strategies without requiring user intervention or manual correction. This self-service capability eliminates the need for users to manually adjust settings or re-issue commands.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS20250006204A1Dynamically adapting fulfillment of a given spoken utterance based on a user that provided the given spoken utterance

Publication Date: 2025.01.02 GOOGLE LLC

AI summary

Implementations described herein relate to determining how to fulfill a spoken utterance based on a user that provided the spoken utterance. For example, implementations can receive a spoken utterance from a user, determine a set of fulfillment actions for the spoken utterance, and determine whether the user that provided the spoken utterance corresponds to a first user or a second user. Further, and in response to determining that the user corresponds to the first user, implementations can select a subset of first fulfillment action(s) from the set, and cause the subset of first fulfillment action(s) to be implemented to satisfy the spoken utterance. Moreover, and in response to determining that the user corresponds to the second user, implementations can select a subset of distinct, second fulfillment action(s) from the set, and cause the subset of second fulfillment action(s) to be implemented to satisfy the spoken utterance.