Hearable Voice Activity Detection Using Ear Canal Acoustic Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless hearables face challenges in detecting vocalizations in noisy environments and ensuring secure voice authentication without bulky and expensive voice accelerometers, which are cumbersome to integrate due to size constraints.
Innovation Solution
Implementing active acoustic sensing through audioplethysmography, which uses acoustic signals to detect vocalizations by forming an acoustic circuit within the ear canal, enabling voice activity detection and multi-factor authentication without additional hardware.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If voice accelerometers are used for voice activity detection and authentication, then detection accuracy and security are improved, but device size increases and cost increases
Solution Approach 1:
The patent replaces the mechanical voice accelerometer sensor with an acoustic signal processing system. The hearable device transmits acoustic signals through the ear canal and analyzes the reflected signals to detect vocal cord vibrations and jaw movements, thereby identifying voice activity without requiring physical acceleration sensors. This substitution eliminates bulky hardware while maintaining detection capability through signal analysis of acoustic reflections.
Solution Approach 2:
The patent introduces acoustic signals as an intermediary medium to detect voice activity. Instead of directly measuring mechanical acceleration, the system uses transmitted acoustic waves that interact with the user's anatomy (ear canal, jaw, vocal cords) and analyzes the modified reflections. These acoustic intermediaries carry information about voice activity while allowing the use of smaller transducers rather than large accelerometer arrays.
2Reliability
If voice accelerometers are used for voice authentication, then security against voice attacks is improved, but device complexity increases
Solution Approach 1:
The patent replaces complex mechanical voice authentication systems with acoustic signal analysis. By transmitting acoustic signals and analyzing their interaction with the user's unique anatomical structures (ear canal geometry, jaw composition, vocal cord characteristics), the system creates a biometric authentication mechanism that is simpler to integrate than traditional voice accelerometers while providing enhanced security against replay attacks through multi-factor verification.
Solution Approach 2:
The patent makes the acoustic transducer serve multiple functions: it acts as both a speaker for transmitting acoustic signals and a microphone for receiving reflected signals. This multi-functionality reduces the number of separate components needed compared to dedicated voice authentication hardware, thereby reducing device complexity while maintaining authentication capability through the same dual-purpose acoustic pathway.
3Measurement precision
If additional hardware is added for voice activity detection, then detection capability in noisy environments is improved, but manufacturing cost increases
Solution Approach 1:
The patent enables existing acoustic transducers in hearable devices to perform voice activity detection in addition to their primary audio playback and recording functions. By processing acoustic reflections from the ear canal and jaw movements, the same transducers provide voice detection capability without requiring separate dedicated sensors, thereby avoiding additional manufacturing costs while enhancing detection capability in noisy environments through multiple signal analysis parameters.
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
Enhances voice activity detection in noisy conditions and provides robust security by directly measuring vocalizations, reducing the need for bulky components and improving user experience with smaller, more affordable hearables.
Implementation Method 1
transmitting, during a first time period, an ultrasound transmit signal that propagates within at least a portion of an ear canal of a user
Implementation Method 2
receiving, during the first time period, an ultrasound receive signal. The ultrasound receive signal represents a version of the ultrasound transmit signal with one or more characteristics modified based on the propagation within the ear canal and based on a vocalization made by the user
Data Source
AI summary
Techniques and apparatuses are described that perform voice activity detection using active acoustic sensing. By transmitting and receiving acoustic signals, a hearable can recognize changes in an acoustic circuit to perform voice activity detection. With active acoustic sensing, the hearable can detect a vocalization made by a user in a noisy and/or loud environment. As such, the hearable can support a voice user interface (VUI) by providing an indication of when the user is speaking. The hearable can also support multi-factor voice authentication to enhance security and provide robust protection from voice attacks. In addition to being relatively unobtrusive, some hearables can be configured to support voice activity detection using active acoustic sensing without the need for additional hardware.