Hardware Configuration
Detailed information about the hardware setup used for the TAPS dataset collection, including microphone specifications and data acquisition systems.
Overview
The TAPS dataset collection system consists of three main components: a throat microphone, an acoustic microphone, and a microcontroller unit (MCU). This setup enables simultaneous recording of both throat vibrations and acoustic speech signals.
Hardware Components
Throat Microphone
MEMS accelerometer-based throat microphone designed for accurate vibration measurement.
Specifications
- Model: TDK IIM-42652
- Sampling Rate: 8 kHz
- Resolution: 16-bit
- Dynamic Range: ±4g
- Interface: SPI
Physical Properties
- Radius: 9 mm
- Thickness: 1.7 mm
- Weight: 760.2 mg
- Board Type: FR-4 PCB
Acoustic Microphone
High-quality acoustic microphone for reference speech recording.
Specifications
- Model: CUI Devices CMM-4030D-261
- Sampling Rate: 16 kHz
- Resolution: 24-bit
- Interface: I2S
Microcontroller Unit (MCU)
Central processing unit for data acquisition and synchronization.
Specifications
- Model: STMicroelectronics STM32F301C8T6TR
- Communication: RS232
- Signal Delay: 0.87 ms
Recording Setup
-
Throat Microphone Placement:
Positioned on the supraglottic area of the neck to capture vocal cord vibrations along with essential speech formants.
-
Acoustic Microphone Position:
Placed 30 cm in front of the speaker's face with a nylon filter to prevent pop noise.
-
Environmental Controls:
Recordings conducted in a semi-soundproof room with reflection filters to minimize ambient noise.
-
Power Supply:
DC battery powered to avoid 60 Hz hum noise from power lines.
Technical Considerations
When working with this hardware setup, consider the following:
- Ensure proper throat microphone contact with the skin
- Monitor battery levels to maintain consistent recording quality
- Regular calibration of both microphones
- Verification of synchronization between devices