Interactive Speech Recognition System for Vehicles

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing speech recognition systems in vehicle environments face challenges in accurately recognizing spoken words due to time variations and orthographic or phonetic similarities, failing to consider the overall content of the speech and distinguish between similar words.

Innovation Solution

An interactive speech recognition system that includes a database, a processing unit, and a recognition circuit, which populates a list memory with reference terms and processes spoken phrases to match terms, forming paths by extending previous paths, thereby minimizing user interaction and enhancing accuracy by considering phonetic and orthographic similarities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional speech recognition systems are used, then the system structure is simple, but the recognition accuracy is low due to time variations and orthographic/phonetic similarities

Engineering Contradiction:
Improverecognition accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the speech recognition process into multiple stages: acoustic signal processing, phoneme extraction, word hypothesis generation, and path formation through category-based term matching. This segmentation allows the system to handle complexity systematically, addressing time variations and phonetic similarities through staged processing rather than a single monolithic approach.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a hierarchical category dimension to the recognition process, organizing terms into categories (e.g., nouns, verbs, modifiers) and forming paths through multiple categories. This dimensional approach enables the system to distinguish between orthographically or phonetically similar words by considering their categorical relationships and contextual paths, thereby improving recognition accuracy without overwhelming complexity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If the system considers overall speech content and phonetic similarities, then recognition accuracy improves, but processing time increases

Engineering Contradiction:
Improverecognition accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary actions by pre-defining term categories and preparing reference terms organized by category before actual speech recognition occurs. During speech processing, the system can quickly match spoken terms against pre-organized categorical references and extend paths efficiently, reducing real-time processing time while maintaining high accuracy through comprehensive categorical consideration.

Inventive Principle:
Principle #10Preliminary action

3Ease of operation

If the system uses hierarchical category-based path formation, then user interaction requirements are reduced, but the device complexity increases

Engineering Contradiction:
Improveuser interaction requirementsVSAvoidrecognition circuit complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent implements self-service through automated path formation and term matching based on hierarchical categories. The recognition circuit automatically extends paths by matching terms across categories without requiring user intervention to resolve ambiguities. This automation reduces user interaction requirements while the systematic categorical framework manages the complexity of the recognition circuit in an organized manner.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS9805722B2Interactive speech recognition system
Publication Date: 2017.10.31 CERENCE OPERATING CO
  • US9805722B2 patent drawing
  • US9805722B2 patent drawing

AI summary

An interactive speech recognition system includes a database containing a plurality of reference terms, a list memory that receives the reference terms of category “n,” a processing circuit that populates the list memory with the reference terms corresponding to the category “n,” and a recognition circuit that processes the reference terms and terms of a spoken phrase. The recognition circuit determines if a reference term of category “n” matches a term of the spoken phrase.