Vehicle Speech System Adaptation via User Pace Modeling

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Vehicle speech systems lack adaptability to individual user communication styles and contextual conditions, leading to suboptimal speech recognition and dialog management.

Innovation Solution

A method and system that receive speech data to determine speech pace, create a user model, and generate adaptation parameters for the speech recognition system and dialog manager to adjust parameters such as recording windows, dialog pace, and recognition methods based on the user model and context.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If a generic dialog is used for all users, then the system complexity is reduced, but the adaptability to individual user communication styles deteriorates

Engineering Contradiction:

Improvesystem complexityVSAvoidadaptability to user communication styles

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The system performs preliminary analysis of user speech patterns during a learning phase before actual dialog interactions. Speech pace metrics are calculated and stored in advance, allowing the system to adapt to individual users without adding complexity during real-time operations

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system changes dialog parameters such as response timing, prompt spacing, and interaction pace based on measured speech pace metrics. By dynamically adjusting these parameters according to user-specific characteristics, the system achieves adaptability without requiring a completely custom dialog for each user

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If the speech system is adapted to individual users, then the recognition accuracy and user experience improve, but the device complexity increases

Engineering Contradiction:

Improverecognition accuracyVSAvoiddevice complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The adaptation system is segmented into distinct functional modules: speech pace analysis module, user model creation module, and dialog parameter adjustment module. This modular approach allows the system to achieve high recognition accuracy through specialized processing while managing complexity through clear separation of concerns

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system implements feedback loops where speech pace metrics are continuously measured during user interactions, used to refine user models, and subsequently applied to adjust dialog parameters. This automated feedback mechanism improves recognition accuracy without requiring manual system configuration or increased complexity

Inventive Principle:

Principle #23Feedback

3Ease of operation

If the dialog pace is fixed for all users, then the system operation is simplified, but the ease of operation for individual users deteriorates

Engineering Contradiction:

Improveease of operation for usersVSAvoiddialog management complexity

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The dialog pace transitions from a static fixed value to a dynamic parameter that automatically adjusts based on real-time measurement of user speech characteristics. The system monitors speech pace metrics and dynamically modifies dialog timing and response intervals to match individual user preferences, significantly improving ease of operation

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS9858920B2Adaptation methods and systems for speech systems

Publication Date: 2018.01.02 GM GLOBAL TECHNOLOGY OPERATIONS LLC

AI summary

Adaptation methods and systems are provided for a speech system of a vehicle. In one embodiment a method comprises: receiving speech data; determining a speech pace based on the speech data; determining a user model based on the speech pace; and generating adaptation parameters for at least one of a speech recognition system and a dialog manager based on the user model.