Multi-Band Piezoelectric Voice Accelerometer for Bone-Conducted Sound
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing piezoelectric MEMS voice accelerometers (VAs) face challenges with high noise floors and resonance peaks located outside the bone-conducted voice vibration range, leading to increased pickup of unwanted external noise and distortion of audio signals.
Innovation Solution
A multi-band piezoelectric MEMS voice accelerometer with multiple sensing elements, each having a distinct resonance frequency within the 100 Hz-1 kHz voice vibration range, to enhance sensitivity and reduce noise interference.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a traditional piezoelectric MEMS voice accelerometer is designed with a single resonance peak, then the device structure is simple, but the resonance peak is located outside the bone-conducted voice vibration range causing increased noise pickup and signal distortion
Solution Approach 1:
The single sensing element is divided into multiple sensing elements, each with its own resonance peak tuned to a specific frequency band within the 100 Hz-1 kHz voice vibration range. This segmentation allows the device to cover the entire voice frequency range with multiple specialized sensors rather than one general-purpose sensor, improving signal-to-noise ratio while maintaining focused functionality in each band.
Solution Approach 2:
The resonance frequency parameter of each sensing element is specifically adjusted and tuned to fall within the bone-conducted voice vibration range (100 Hz-1 kHz). By changing the resonance frequency parameter from outside this range to inside this range, the device achieves better alignment with the target signal frequencies, reducing noise pickup and signal distortion.
2Measurement precision
If multiple sensing elements with distinct resonance frequencies are used, then sensitivity within the voice vibration range is enhanced, but the device complexity increases
Solution Approach 1:
The frequency measurement range is segmented into multiple bands, with each sensing element responsible for a specific frequency band. This segmentation enhances measurement precision within each band by dedicating specialized sensing elements to specific frequency ranges, rather than using a single element that must cover the entire range with reduced effectiveness.
Solution Approach 2:
Multiple sensing elements with different resonance frequencies work together as a unified multi-functional system that can detect vibrations across the entire 100 Hz-1 kHz voice vibration range. Each element contributes its specialized sensitivity to a particular frequency band, creating a comprehensive measurement capability that exceeds what a single element could achieve.
3Reliability
If resonance peaks are aligned with the voice vibration band, then distortion is reduced, but the manufacturing precision requirements increase
Solution Approach 1:
The resonance frequency parameter of each sensing element is deliberately changed and tuned to align with specific portions of the voice vibration band. This parameter adjustment, while requiring manufacturing precision, ensures that the resonance peaks occur at frequencies where they are most beneficial for voice detection, thereby reducing distortion and improving overall audio signal quality.
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 multi-band design improves signal-to-noise ratio and reduces distortion by aligning resonance peaks with the voice vibration band, effectively capturing bone-conducted sound while minimizing external noise interference.
Implementation Method 1
piezoelectric voice accelerometers (VAs), which may be used for certain functionality such as to implement bone conduction microphones (BCMs) based on sensing bone-conducted vibrations of the vocal cords
Implementation Method 2
a back cavity enclosed by the plurality of sensing elements and the substrate, wherein a volume of the back cavity extends between the plurality of sensing elements and the substrate, and wherein each respective sensing element of the plurality of sensing elements is configured to vibrate in response to: a first force corresponding to a bone-conducted sound wave coupled into the voice accelerometer; and a second force corresponding to a back cavity pressure coupling between the plurality of sensing elements
Data Source
AI summary
Systems and techniques are provided for detecting bone-conducted sound. A voice accelerometer can include a substrate and a plurality of sensing elements associated with a plurality of frequency bands. Each frequency band can be associated with one or more sensing elements of the plurality of sensing elements having a respective resonance frequency within the frequency band. The voice accelerometer can include a back cavity enclosed by the plurality of sensing elements and the substrate. Each respective sensing element of the plurality of sensing elements can be configured to vibrate in response to a first force corresponding to a bone-conducted sound wave coupled into the voice accelerometer, and a second force corresponding to a back cavity pressure coupling between the plurality of sensing elements, the back cavity pressure coupling based on respective vibration of each sensing element of the plurality of sensing elements.