Secondary-Language Proficiency Modeling for Adaptive Speech Recognition

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Users interacting with automated assistants face challenges in having their secondary language utterances recognized and learning resources tailored to their proficiency level, leading to misinterpretation and inefficient use of language learning tools.

Innovation Solution

A secondary language proficiency measure is determined based on past user interactions to automatically set parameters for language learning applications and bias speech recognition, rendering language output adaptively to the user's proficiency level.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If an automatic speech recognition model is configured to recognize utterances in the primary language of the user, then recognition accuracy for primary language is improved, but recognition capability for secondary language deteriorates

Engineering Contradiction:

Improvespeech recognition accuracyVSAvoidlanguage recognition capability

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The speech recognition system dynamically adjusts the language model based on detected user intent and context. When a secondary language is detected, the system switches to an appropriate language model for that language, allowing the recognition capability to adapt between primary and secondary languages rather than being fixed to one language configuration

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes the language parameter of the speech recognition model based on the detected language of the utterance. By monitoring language-specific features in the input signal and changing the active language model parameter, the system maintains high recognition accuracy across multiple languages without requiring simultaneous configuration of all language models

Inventive Principle:

Principle #35Parameter changes

2Quantity of substance

If language learning resources are provided without considering user proficiency level, then resource availability is improved, but learning effectiveness deteriorates

Engineering Contradiction:

Improvelanguage learning resourcesVSAvoidlearning effectiveness

Core Design Contradiction:

Quantity of substanceVSReliability

Solution Approach 1:

The system applies different quality levels of language learning resources to different user proficiency levels. Beginners receive simplified content with more support features, while advanced users receive more complex materials. This local differentiation of resource quality based on user capability ensures that each user receives appropriately matched resources rather than a one-size-fits-all approach

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system performs preliminary assessment of user proficiency level before providing language learning resources. By evaluating the user's current language ability in advance and configuring the resource delivery parameters accordingly, the system ensures that resources are appropriately matched to user needs from the start, preventing the provision of resources that would be either too difficult or too simplistic

Inventive Principle:

Principle #10Preliminary action

3Measurement precision

If manual configuration of language learning parameters is required, then parameter accuracy is improved, but user interaction time increases

Engineering Contradiction:

Improveparameter accuracyVSAvoidsetup time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs self-service by automatically detecting user language proficiency level and configuring appropriate parameters without requiring manual user input. The system monitors user interactions, analyzes language usage patterns, and autonomously adjusts language learning parameters, thereby eliminating setup time while maintaining parameter accuracy through continuous adaptive monitoring

Inventive Principle:

Principle #25Self-service

4Adaptability or versatility

If speech recognition is performed without biasing toward user proficiency level, then recognition coverage is improved, but recognition accuracy for user's level deteriorates

Engineering Contradiction:

Improverecognition coverageVSAvoidrecognition accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The system changes the parameter of language model selection based on detected user proficiency level. When processing speech input, the system adjusts which language model is activated - using beginner-appropriate models for low-proficiency users and advanced models for high-proficiency users - thereby maintaining high recognition accuracy for the user's actual level while still covering a broad range of possible inputs through multiple configured models

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS12412570B2Determining and utilizing secondary language proficiency measure

Publication Date: 2025.09.09 GOOGLE LLC

AI summary

Implementations relate to determining a secondary language proficiency measure, for a user in a secondary language (i.e., a language other than a primary language specified for the user), where determining the secondary language proficiency measure is based on past interactions of the user that are related to the secondary language. Those implementations further relate to utilizing the determined secondary language proficiency measure to increase efficiency of user interaction(s), such as interaction(s) with a language learning application and/or an automated assistant. Some of those implementations utilize the secondary language proficiency measure in automatically setting value(s), biasing automatic speech recognition, and/or determining how to render natural language output.