Voice Quality Evaluation via Envelope Spectrum Extraction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current signal-domain-based voice quality evaluation methods have high computational complexity and resource consumption, making them inefficient for monitoring large and complex voice communications networks.

Innovation Solution

A low-complexity voice quality evaluation method that directly obtains the time envelope of a voice signal, performs time-to-frequency conversion, extracts feature parameters, calculates voice quality parameters using a network evaluation model, and combines these parameters for comprehensive analysis, avoiding the simulation of human auditory perception with high-complexity cochlea filters.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a cochlea filter is used to simulate human auditory system for voice quality evaluation, then measurement precision is improved, but device complexity and computational complexity increase significantly

Engineering Contradiction:

Improvevoice quality evaluation accuracyVSAvoidcomputational complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts only the essential feature (envelope spectrum) from the voice signal, omitting the complex cochlea filter simulation. By taking out the critical component (envelope extraction) while discarding the unnecessary complexity (cochlea filter), the system achieves voice quality evaluation without the heavy computational burden of full auditory model simulation.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent replaces the expensive and complex cochlea filter simulation with a simpler, more efficient envelope spectrum extraction method. This substitution uses computationally lightweight operations (FFT, envelope detection) instead of resource-intensive cochlear modeling, making the system more suitable for network monitoring applications.

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

2Measurement precision

If cochlea filter bank is used to divide spectrum band for processing, then measurement precision is improved, but use of energy and computational complexity increase

Engineering Contradiction:

Improvevoice quality evaluation accuracyVSAvoidresource consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent extracts the envelope spectrum directly from the voice signal using efficient signal processing techniques, avoiding the need to process the entire spectrum through multiple cochlear filters. This extraction approach captures the essential voice quality information with significantly lower energy consumption than full spectral analysis via cochlea filter banks.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of performing complete spectral analysis across all frequency bands using cochlear filters, the patent applies partial action by focusing only on envelope extraction and key frequency band analysis. This selective processing achieves sufficient voice quality evaluation accuracy while reducing overall computational energy consumption.

Inventive Principle:

Principle #16Partial or excessive action

3Measurement precision

If complex convolution operation processing is performed on voice signal in each key frequency band, then measurement precision is improved, but productivity decreases due to high computational complexity

Engineering Contradiction:

Improvevoice quality evaluation accuracyVSAvoidmonitoring efficiency

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces the mechanical convolution operations required for cochlear filter processing with more efficient signal processing techniques. By substituting complex time-domain convolution with frequency-domain envelope extraction and spectral analysis, the system achieves comparable measurement precision with much higher processing speed and productivity.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent segments the voice signal processing into distinct functional stages: envelope extraction, envelope spectrum calculation, and voice quality parameter derivation. This segmentation allows each stage to be optimized independently, improving overall processing efficiency while maintaining measurement precision through targeted analysis of critical signal features.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS10497383B2Voice quality evaluation method, apparatus, and device

Publication Date: 2019.12.03 HUAWEI TECH CO LTD

AI summary

A voice quality evaluation method includes obtaining a time envelope of a voice signal. The method includes performing time-to-frequency conversion on the time envelope to obtain an envelope spectrum. The method includes performing feature extraction on the envelope spectrum to obtain a feature parameter. The method includes performing voice quality evaluation in voice communications according to the feature parameter to obtain a first voice quality parameter of the voice signal. The method includes calculating a second voice quality parameter of the voice signal by using a network parameter evaluation model. The method includes performing a comprehensive analysis according to the first voice quality parameter and the second voice quality parameter to obtain a quality evaluation parameter of the voice signal that is input in the band.