Hands-Free Double Talk Evaluation Using ASR and CVC Sounds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing evaluation methods for hands-free communication systems do not adequately consider subjective quality, often rating systems as deficient despite achieving good perceived quality, and may not accurately measure the subjective perceptual quality due to reliance on objective measures that do not correspond to human experience.
Innovation Solution
A system utilizing an automatic speech recognition engine to evaluate the quality of hands-free communication systems by selecting consonant-vowel-consonant (CVC) or vowel-consonant-vowel (VCV) sounds from a database and transmitting them to assess the system's subjective perceptual quality, allowing for remote evaluation and minimizing human intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If objective measures are used to evaluate hands-free systems, then measurement precision is improved, but the evaluation does not correspond to subjective quality
Solution Approach 1:
The patent introduces an automatic speech recognition engine as an intermediary between the hands-free system under test and the evaluation process. This intermediary translates the system's output into recognizable speech patterns, allowing objective measurement while maintaining correspondence with subjective quality perception. The ASR engine acts as a mediator that bridges the gap between technical measurements and human perceptual quality.
2Device complexity
If artificial speech signals are used to model real speech, then device complexity is reduced, but the evaluation may not measure subjective perceptual quality
Solution Approach 1:
The patent uses artificial speech signals as simplified copies of real speech, generated through concatenation of phoneme segments. These synthetic speech samples replicate the essential characteristics of natural speech without requiring actual human speakers, thereby reducing complexity while maintaining sufficient fidelity for evaluating hands-free system performance in double-talk scenarios.
3Measurement precision
If switching times and attenuation levels are measured, then measurement precision is improved, but the subjective perceptual quality is not measured
Solution Approach 1:
The patent replaces direct measurement of physical parameters (switching times, attenuation levels) with a cognitive-based evaluation system using automatic speech recognition. Instead of measuring mechanical/electrical characteristics, the system uses linguistic recognition to assess whether the hands-free system successfully transmits intelligible speech, which directly correlates with subjective perceptual quality.
Data Source
AI summary
A system evaluates a hands free communication system. The system automatically selects a consonant-vowel-consonant (CVC), vowel-consonant-vowel (VCV), or other combination of sounds from an intelligent database. The selection is transmitted with another communication stream that temporally overlaps the selection. The quality of the communication system is evaluated through an automatic speech recognition engine. The evaluation occurs at a location remote from the transmitted selection.