Head-Mounted RF Sensing for Continuous Silent Speech Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing silent speech detection technologies are invasive or require components inside the oral cavity, limiting user comfort and practicality.

Innovation Solution

A non-invasive method using radio frequency (RF) sensing devices coupled to the head, tracking speech articulator movements to detect and translate non-audible speech, with antennas positioned on the head to maximize signal transmission and minimize environmental noise interference.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If invasive measuring techniques (PMA, EMA, EPG, EOS) are used to detect silent speech, then measurement precision is improved, but ease of operation deteriorates due to requiring components inside the oral cavity

Engineering Contradiction:

Improvespeech detection accuracyVSAvoiduser comfort

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The patent replaces invasive mechanical/electrical measurement systems (electrodes, magnets, coils inside oral cavity) with a non-invasive electromagnetic sensing system using RF antennas that detect speech articulator movements through the head externally, eliminating the need for internal components while maintaining detection capability

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces RF electromagnetic waves as an intermediary medium to transmit information about speech articulator movements from the internal location to external sensors, allowing indirect measurement without physical intrusion into the oral cavity

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If RF sensing devices with antennas are positioned on the head to detect speech articulator movements, then measurement precision is improved, but object-generated harmful factors worsen due to environmental noise interference

Engineering Contradiction:

Improvespeech signal detection accuracyVSAvoidenvironmental noise interference

Core Design Contradiction:

Measurement precisionVSObject-generated harmful factors

Solution Approach 1:

The patent applies local quality by directing RF energy specifically at speech articulators using targeted antenna positioning and beamforming techniques, concentrating the sensing field where needed while minimizing exposure to environmental noise in surrounding areas

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent converts the potentially harmful effect of environmental RF noise into a beneficial signal by using the same RF sensing mechanism to detect speech articulator movements, where the modulated RF signals containing speech information are extracted from the noise background through signal processing

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

3Ease of operation

If non-invasive RF sensing is used to detect silent speech, then ease of operation is improved by eliminating physical contact with the face, but measurement precision deteriorates due to signal transmission challenges

Engineering Contradiction:

Improveuser comfortVSAvoidspeech detection accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent applies preliminary action by performing calibration procedures before actual speech detection, where the system learns and stores the user's anatomical characteristics and speech patterns in advance, enabling more accurate real-time detection without physical contact

Inventive Principle:

Principle #10Preliminary action

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

Enables real-time, accurate translation of non-audible speech with a word error rate less than 50% and minimal delay, allowing continuous speech without physical contact with the face, enhancing user comfort and communication efficiency.

Implementation Method 1

using a radio frequency (RF) sensing device coupled to a head of a user to collect the data associated with the non-audible speech of the user

Methodology Applied

Scientific EffectRadio frequency signal transmission and detection: Radar

Implementation Method 2

the data associated with the non-audible speech of the user comprises RF signal data associated with movement of one or more speech articulators of the user

Methodology Applied

Scientific EffectDoppler effect: Doppler Effect

Data Source

PatentUS12469488B2Methods for non-audible speech detection

Publication Date: 2025.11.11 REFLEX TECHNOLOGIES INC

AI summary

Provided herein is a method for non-audible speech detection and output. The method comprises providing a radio frequency (RF) sensing device configured to be coupled to a head of a user. The method further comprises using the RF sensing device to collect RF signal data associated with movement of one or more speech articulators of the user. The method further comprises outputting or facilitating an output comprising a non-audible speech translation using at least in part processed RF signal data, wherein the non-audible speech of the user comprises continuous speech by the user.