Speech Detection via Time-Domain Amplitude Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current speech detection algorithms are either inaccurate or require complex analysis, being sensitive to audio quality changes and computational intensive, limiting their implementation in systems with limited computing power, and fail to quantify the amount of speech in an audio signal.

Innovation Solution

A method that calculates the audio signal speech grade by analyzing the dynamic behavior and ratio of high and low volume parts, using segment and block values to determine the presence and quantity of speech, providing a computationally lightweight solution that is amplitude and language independent.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If complex frequency analysis or phonetic comparison is used for speech detection, then measurement precision is improved, but use of energy increases and device complexity increases

Engineering Contradiction:

Improvespeech detection accuracyVSAvoidcomputational power consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent extracts only the essential dynamic characteristics (zero-crossing rate, peak detection, RMS variations) from the audio signal that are sufficient for speech detection, rather than performing complete frequency analysis or phonetic comparison. This selective extraction of critical features maintains detection accuracy while dramatically reducing computational requirements and energy consumption.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of analyzing the audio signal in the frequency domain through complex transforms, the patent inverts the approach by working directly in the time domain with simple amplitude-based metrics. This inversion from frequency-domain complex analysis to time-domain simple measurement achieves comparable speech detection performance with minimal computational overhead.

Inventive Principle:

Principle #13The other way round (Inversion)

2Measurement precision

If complex frequency analysis or phonetic comparison is used for speech detection, then measurement precision is improved, but device complexity increases

Engineering Contradiction:

Improvespeech detection accuracyVSAvoidalgorithm complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

Inventive Principle:

Principle #13The other way round (Inversion)

3Use of energy by moving object

If simple speech detection algorithms are used, then use of energy is reduced, but measurement precision deteriorates

Engineering Contradiction:

Improvecomputational power consumptionVSAvoidspeech detection accuracy

Core Design Contradiction:

Use of energy by moving objectVSMeasurement precision

Solution Approach 1:

The patent creates simplified copies of speech signal characteristics by computing basic time-domain features (zero-crossing rate, peak amplitude, RMS) that replicate the essential dynamic behavior of speech. These simplified proxies capture speech presence accurately enough for detection purposes while requiring minimal computational resources compared to full spectral analysis.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent changes the measurement parameters from complex frequency-domain representations to simple time-domain amplitude characteristics. By monitoring how these basic parameters change over time (dynamic behavior), the system achieves speech detection accuracy comparable to complex algorithms while consuming minimal computational power.

Inventive Principle:

Principle #35Parameter changes

4Device complexity

If simple speech detection algorithms are used, then device complexity is reduced, but measurement precision deteriorates

Engineering Contradiction:

Improvealgorithm complexityVSAvoidspeech detection accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

Inventive Principle:

Principle #26Copying

Solution Approach 2:

Inventive Principle:

Principle #35Parameter changes

5Loss of time

If speech detection is not performed, then processing time is reduced, but loss of information increases due to recording noise or silence unknowingly

Engineering Contradiction:

Improveprocessing timeVSAvoidcompliance information

Core Design Contradiction:

Loss of timeVSLoss of information

Solution Approach 1:

The patent performs preliminary speech detection analysis using simple time-domain features before committing to full audio processing or recording. This preliminary check quickly identifies whether speech is present, preventing wasteful processing of silent or noisy segments while ensuring compliance-critical speech segments are captured and processed appropriately.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Inventive Principle:

Principle #26Copying

Data Source

PatentUS9916846B2Method and system for speech detection

Publication Date: 2018.03.13 NICE LTD

AI summary

A system and method for determining an amount of speech in an audio signal may include for example: obtaining segments of the audio signal, wherein the segments are grouped into blocks; for each one of the segments, calculating a segment value indicative of an amplitude of the audio signal of a respective segment; for each one of the blocks calculating a block value indicative of the amplitude of the audio signal of a respective block; and calculating an audio signal speech grade based on segment values and block values, wherein the audio signal speech grade is indicative of the amount of speech in the audio signal.