Contextual Resolver for Ambiguous Voice Requests

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional voice-controlled devices struggle with processing ambiguous voice requests due to their reliance on explicit information, leading to frustrating user experiences as they often require additional input to clarify intentions, especially when implicit context is lacking.

Innovation Solution

The implementation of a contextual resolver engine that utilizes machine learning algorithms and natural language processing to identify and incorporate contextual information such as relational, location, and time-based contexts, allowing voice-controlled devices to interpret user requests more accurately and efficiently by filling in missing information.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional voice-controlled devices rely on explicit information from voice commands, then the processing logic is simple and device complexity is low, but the ability to handle ambiguous requests is poor and user experience deteriorates

Engineering Contradiction:

Improveability to handle ambiguous requestsVSAvoidprocessing logic complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent introduces a contextual resolver as an intermediary component between the voice command receiver and the command executor. This contextual resolver analyzes contextual information (location, time, device state) and resolves ambiguous references by matching them with relevant context, thereby improving reliability without requiring fundamental changes to the core voice processing architecture.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system performs preliminary actions by collecting and storing contextual information (location data, time stamps, device state) before the voice command is processed. This pre-collected context is then available to resolve ambiguities, allowing the system to handle ambiguous requests reliably without adding complexity to the real-time command processing logic.

Inventive Principle:

Principle #10Preliminary action

2Measurement precision

If additional contextual processing is added to resolve ambiguous voice requests, then the accuracy of interpreting user intent improves, but the number of processing steps increases and efficiency decreases

Engineering Contradiction:

Improveaccuracy of interpreting user intentVSAvoidprocessing efficiency

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The contextual resolver applies partial processing by selectively analyzing only the contextual information that is relevant to resolving the specific ambiguity in the current voice command. Rather than processing all possible context data, the system performs just enough contextual analysis to resolve the ambiguity, thereby maintaining high accuracy while preserving processing efficiency.

Inventive Principle:

Principle #16Partial or excessive action

3Manufacturing precision

If the system requires additional explicit input from users to clarify ambiguous requests, then the precision of command execution is high, but the ease of operation decreases and user experience worsens

Engineering Contradiction:

Improveprecision of command executionVSAvoiduser interaction simplicity

Core Design Contradiction:

Manufacturing precisionVSEase of operation

Solution Approach 1:

The system performs self-service by automatically resolving ambiguous references using available contextual information without requiring additional user input. The contextual resolver autonomously matches ambiguous terms with relevant context (e.g., resolving 'this song' to the currently playing track), thereby maintaining precision while significantly improving ease of operation.

Inventive Principle:

Principle #25Self-service

4Adaptability or versatility

If the system processes only explicit information from voice commands, then the device complexity is low and processing is fast, but the adaptability to handle implicit context is poor

Engineering Contradiction:

Improveability to process implicit contextVSAvoidcontextual processing capability

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The contextual resolver is designed as a universal component that can handle multiple types of contextual information (location, time, device state, currently playing media) and apply them to resolve various kinds of ambiguities. This multi-functional approach enables the system to process implicit context across different scenarios without requiring separate specialized modules for each context type.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS11756541B1Contextual resolver for voice requests

Publication Date: 2023.09.12 AMAZON TECH INC

AI summary

Contextual data associated with an ambiguous voice command to a voice-controlled device can be used to provide information to complete the voice request without requiring additional input or an explicit statement of information from the user. Contextual data and voice request data are inputs to a machine learning ranker that produces a confidence score useful for selecting a data object and providing implicit information to fulfill the voice request. Once the data object is selected, the voice request is performed by the voice-controlled device.