Flexible Piezoelectric Voice Sensor with Frequency Separation Channels

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing voice recognition systems for IoT devices consume high power due to the use of microphones, ADCs, and DSPs, limiting continuous operation and accuracy, especially in mobile and wearable applications, and struggle to recognize natural conversational tones effectively.

Innovation Solution

A flexible piezoelectric voice recognition sensor using a single flexible thin film with a piezoelectric material layer and electrode channels of varying lengths, arranged in a trapezoidal shape, separates voice signals by frequency and converts mechanical vibrations into electric signals, simplifying the voice recognition circuit and reducing power consumption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a combination of microphone, ADC and DSP is used for voice recognition, then voice recognition accuracy is improved, but power consumption increases significantly

Engineering Contradiction:

Improvevoice recognition accuracyVSAvoidpower consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The voice recognition system is segmented into multiple frequency separation channels (first channel, second channel, third channel, etc.), each handling specific frequency ranges. This segmentation allows parallel processing of different frequency components, improving recognition accuracy while distributing computational load to reduce power consumption in each individual channel.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent replaces the traditional mechanical/electronic system (microphone + ADC + DSP) with a piezoelectric-based system. The piezoelectric element directly converts acoustic vibrations into electrical signals with frequency separation, eliminating the need for separate ADC and DSP components, thereby significantly reducing power consumption while maintaining voice recognition accuracy.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Speed

If voice recognition is always in standby state, then response speed is improved, but power consumption increases

Engineering Contradiction:

Improveresponse speedVSAvoidpower consumption

Core Design Contradiction:

SpeedVSUse of energy by moving object

Solution Approach 1:

The piezoelectric element operates periodically, converting acoustic vibrations into electrical signals only when voice input is detected. The frequency separation channels process signals in a periodic manner, activating only when needed, thus achieving fast response to voice commands while consuming minimal power during idle periods compared to continuous operation of traditional systems.

Inventive Principle:

Principle #19Periodic action

3Adaptability or versatility

If traditional voice recognition system is used, then voice recognition function is achieved, but device complexity increases

Engineering Contradiction:

Improvevoice recognition functionVSAvoidcircuit complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent merges the functions of microphone, frequency separation, ADC, and DSP into a single integrated piezoelectric device with multiple frequency separation channels. This consolidation eliminates the need for separate components and complex interconnections, significantly reducing device complexity while maintaining complete voice recognition functionality.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The piezoelectric element serves multiple functions simultaneously: it acts as the acoustic sensor, performs frequency separation through its channel structure, converts mechanical vibrations to electrical signals, and provides the basis for digital processing. This multi-functionality eliminates the need for separate dedicated components, reducing overall system complexity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The solution enables ultra-low power voice recognition with reduced power consumption, allowing continuous operation and accurate recognition of voice signals in IoT devices without the need for microphones, ADCs, and DSPs, enhancing sensitivity and speed while eliminating the requirement for a start button.

Implementation Method 1

converting the separated voice signal from a mechanical vibration signal into an electric signal

Methodology Applied

Scientific EffectPiezoelectric effect: Piezoelectric Effect

Data Source

PatentUS10436631B2Ultra-low power flexible piezoelectric audio recognition sensor for internet of things

Publication Date: 2019.10.08 KOREA ADVANCED INST OF SCI & TECH

AI summary

Disclosed is a piezoelectric voice recognition sensor, which includes a flexible thin film, a piezoelectric material layer laminated on the flexible thin film, and an electrode laminated on the piezoelectric material layer, wherein the electrode includes a plurality of frequency separation channels arranged in a row, and the plurality of frequency separation channels have different lengths from each other. The piezoelectric voice recognition sensor separates a voice, recognized using a plurality of frequency separation channels having a trapezoidal shape, through the plurality of channels depending on frequencies, and simultaneously converts the separated voice signals from mechanical vibration signals into electric signals by means of the flexible piezoelectric element so that the converted electric signals are recognized.