Bone and Air Conduction Microphones for Speech Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional head-mounted wearable devices (HMWDs) face challenges in accurately determining if the user is speaking due to interference from ambient noise, as air conduction microphones detect sounds from the environment, leading to incorrect determinations.
Innovation Solution
The use of both bone conduction and air conduction microphones to generate voice activity data, where the bone conduction microphone, in contact with the user's skin, primarily detects the user's speech, while the air conduction microphone detects ambient noise, allowing for the processing of both signals to determine the presence of speech and differentiate it from external noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If air conduction microphone is used to detect user speech, then speech detection is possible, but ambient noise interference causes incorrect determinations
Solution Approach 1:
The patent segments the speech detection function into two separate sensing pathways: air conduction microphone for ambient sound capture and bone conduction sensor for direct skull vibration detection. By dividing the detection system, the patent isolates the user's speech signal (detected primarily through bone conduction) from ambient noise (detected primarily through air conduction), thereby resolving the contradiction between speech detection capability and noise interference resistance
Solution Approach 2:
The patent introduces bone conduction as an intermediary pathway for speech signal transmission. Instead of directly capturing air-borne sound waves that mix with ambient noise, the system uses bone vibrations as a mediator to transmit speech signals directly from the user's vocal apparatus to the sensor, bypassing the noisy air conduction path and improving speech detection accuracy
2Reliability
If only air conduction microphone is used, then device structure is simple, but speech detection accuracy deteriorates in noisy environments
Solution Approach 1:
The patent merges two different sensing technologies (air conduction microphone and bone conduction sensor) into a unified speech detection system. By combining these complementary sensing modalities, the patent achieves reliable speech detection in noisy environments while maintaining a relatively compact wearable device structure, as both sensors can be integrated into the same headset form factor
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
This approach effectively filters out ambient noise, enabling accurate detection of the user's speech in noisy environments, enhancing the functionality and user experience of HMWDs by ensuring that only the user's voice triggers device responses.
Implementation Method 1
the bone conduction microphone, in contact with the user's skin, primarily detects the user's speech
Implementation Method 2
the air conduction microphone detects ambient noise
Data Source
AI summary
A head-mounted wearable device incorporates a transducer that operates as a bone conduction (BC) microphone. Vibrations from a user's speech are transferred through the head of the user to the BC microphone. An air conduction (AC) microphone detects sound transferred via air. Signals from the BC microphone and the AC microphone are compared to determine if a common signal is present in both. For example, both signals may have a cross-correlation that exceeds a threshold value. Based on the comparison, voice activity data is generated that indicates the user wearing the device is speaking.