Voice Input Setting Priority Inversion for Intent Accuracy

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

When a user requests a process based on spoken voice, settings in a prohibition relationship can lead to unintended selection, as settings spoken later are prioritized, potentially overriding the user's intended settings.

Innovation Solution

An information processing apparatus that receives a request for a process based on spoken voice and, in cases where the request includes settings in a prohibition relationship, controls the process to use the setting spoken earlier if it precedes the other setting in the voice request.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the system prioritizes settings spoken later in voice requests, then the process is performed with the most recently mentioned setting, but the setting intended by the user may not be selected

Engineering Contradiction:

Improveaccuracy of setting selectionVSAvoidreliability of user intent recognition

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent inverts the conventional approach by prioritizing settings spoken earlier rather than later. When multiple settings are provided in a voice request, the system selects the first setting mentioned instead of the last, which reverses the typical last-in-first-out selection logic and better reflects user intent in natural speech patterns.

Inventive Principle:

Principle #13The other way round (Inversion)

Solution Approach 2:

The patent applies preliminary action by establishing priority based on the temporal order of settings in the voice request. The system processes and prioritizes settings as they are spoken, giving precedence to earlier mentions before finalizing the selection, thus preventing later settings from overriding user intent.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If the system selects settings based on spoken order, then user intent is better captured, but the processing logic becomes more complex compared to screen operation selection

Engineering Contradiction:

Improvereliability of user intent recognitionVSAvoidcomplexity of selection logic

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent changes the selection parameter from temporal recency (last setting spoken) to temporal precedence (first setting spoken). This parameter change simplifies the logic by establishing a clear first-setting-priority rule that is easier to implement and process than complex conflict resolution algorithms, while maintaining high reliability in user intent recognition.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS12205583B2Information processing apparatus and non-transitory computer readable medium storing program

Publication Date: 2025.01.21 FUJIFILM BUSINESS INNOVATION CORP

AI summary

An information processing apparatus includes a processor configured to receive a request for a process based on a spoken voice of a user, and in a case where the request includes a first setting and a second setting that are in a prohibition relationship, perform a control for performing the process with the first setting in a case where the first setting is spoken earlier than the second setting in the spoken voice.