Dynamic Speech Grammar Weight Adjustment

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional speech recognition systems use static grammar weights optimized for theoretical average users, leading to decreased accuracy as actual usage patterns diverge from estimates, with no technology to dynamically adjust weights based on actual user usage.

Innovation Solution

A method to automatically record and dynamically adjust speech grammar weights based on usage data, increasing the relative weight of frequently used words and phrases, ensuring improved recognition accuracy over time.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If static grammar weights optimized for theoretical average users are used, then speech recognition accuracy is maintained for generic users, but accuracy degrades as actual usage patterns diverge from theoretical estimates

Engineering Contradiction:

Improvespeech recognition accuracyVSAvoidadaptability to individual usage patterns

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent transforms static grammar weights into dynamic weights that automatically adjust based on actual usage statistics. The system continuously monitors speech command usage frequencies and recalibrates weights to reflect real-world patterns, enabling the grammar to adapt to individual user behaviors while maintaining recognition accuracy across diverse usage scenarios

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system implements a feedback loop where usage data from actual speech recognition operations is collected, analyzed, and used to adjust grammar weights. This closed-loop approach allows the system to learn from actual user interactions and continuously improve recognition accuracy by reinforcing frequently used commands and adjusting less common ones

Inventive Principle:

Principle #23Feedback

2Reliability

If grammar weights are statically optimized for theoretical average usage, then computational overhead remains low, but recognition accuracy increasingly degrades over time as usage patterns diverge

Engineering Contradiction:

Improvespeech recognition accuracyVSAvoidtime for weight adjustment

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-calculating and storing multiple sets of grammar weights corresponding to different usage scenarios. When usage patterns shift, the system can quickly switch between pre-computed weight sets or apply incremental adjustments based on accumulated usage data, avoiding time-consuming real-time recalculations while maintaining accuracy

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS8131548B2Dynamically adjusting speech grammar weights based on usage

Publication Date: 2012.03.06 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

A speech processing method can automatically and dynamically adjust speech grammar weights at runtime based upon usage data. Each of the speech grammar weights can be associated with an available speech command contained within a speech grammar to which the speech grammar weights apply. The usage data can indicate a relative frequency with which each of the available speech commands is utilized.