[Introduction]TWS (True Wireless Stereo) true wireless Bluetooth headset has extremely high portability compared with traditional headsets. With the addition of new functions such as active noise reduction and spatial audio, the functionality of TWS headphones is more abundant, the user experience is continuously improved, and the rapid popularization of TWS headphones is further promoted.
In daily life, the application scenarios of voice calls are relatively complex. For example, in public places such as subways, offices, and airports, there are high requirements for call effects. In uplink calls, the elimination of environmental noise has become a rigid requirement. How to further improve the noise reduction effect of TWS headsets, so as to highlight the differences of products and improve the competitiveness of products, has become one of the key directions for many manufacturers to deploy products.
Detection principle of bone vibration sensor
When a person speaks, the vibration signal generated by the vocal cords propagates in two ways: one is to propagate outward through the air medium, and the other way is to propagate outward through the bones and muscles of the human brain, causing the vibration of the auricle. . The bone vibration sensor uses the latter transmission method to detect the vibration signal of the auricle and pick up the wearer’s voice information. The bone vibration sensor is not sensitive to the sound wave signal propagating in the air, and has a natural inhibitory effect on the sound signal propagating in the air. Therefore, the call noise reduction algorithm is simpler and more natural, the noise suppression is more effective, and it can provide a better uplink. call effect.
Comparison of noise reduction methods for TWS headphones
There are many types of TWS earphones on the market. For ENC (Environmental Noise Cancellation) noise reduction, according to different price and performance positioning, there are various product forms such as single microphone, dual microphone, triple microphone and bone vibration.
The single microphone solution has an advantage in price, the main control chip requires a lower threshold, and the structure is not limited. The neural network algorithm is used to identify and filter out noise signals. However, the noise reduction effect of the single-microphone solution is general, and the noise recognition and suppression of complex scenes is not very obvious. Dual microphones can be adapted to a variety of forms, have a better experience in most high-noise scenarios, and are cost-effective, and are currently the mainstream solution for ENC noise reduction.
Dual microphone ENC noise reduction, using two microphones with a certain distance, so that the sound signals in different directions can be recognized. The minimum distance between the two microphones is generally 10mm. The algorithm uses a relatively mature beam forming algorithm, the waveform points to the sound source of the wearer’s speech, enhances the signal sensitivity in the direction of the sound source, and suppresses the signal sensitivity in other directions, thereby eliminating noise.
Sanmai ENC noise reduction generally reuses the ANC active noise reduction feedback microphone (Feedback Microphone, FB). The FB microphone is used to detect the sound signal transmitted by the vocal cords to the earphone through the cochlea. Most of them are in-ear wearing methods. The rubber sleeve of the earphone has a good sealing effect, which can isolate the external environmental noise. Sanmai ENC noise reduction, suitable for most scenarios, has good wind noise resistance, but the algorithm is complex and the price is relatively high.
The combination of microphone and bone vibration has better anti-interference performance of environmental human voice, and the distortion of the caller’s voice is small. Wind noise is a pain point for most headphones. When the wind speed is greater than 5m/s, the microphone will have saturation distortion. At this time, the microphone loses the ability to capture voice signals, and the bone vibration sensor does not respond to air fluctuations, so even in the case of high wind speed , still can accurately capture the voice signal. In addition, for bean-type headphones, due to the compact space and small size, it is difficult for the two microphones to be separated, so the noise reduction effect of the dual-mic ENC is not good in this case. A call microphone and a bone vibration sensor can achieve a better call effect, strong wind noise resistance, and good noise suppression effect. The algorithm does not require complex beamforming, and a simple voice algorithm can be implemented. The overall solution Cost-effective, used by more and more headphone manufacturers.
Device Design of Bone Vibration Sensor
Based on MEMS micromachining technology, a micron-sized elastic beam structure and a mass block of a certain weight are processed on a silicon substrate to form a vibration-sensitive mechanical structure. The capacitor plate on the mass block and the capacitor plate on the substrate form a comb-shaped movable capacitor. When vibration is applied to the device, the mass block vibrates, which drives the capacitor comb teeth to vibrate. change, so the capacitance changes.
The ASIC chip inside the device extracts, converts and amplifies the weak capacitive signal, and outputs the vibration signal in the form of voltage, which is similar to the output signal of the traditional silicon microphone, which is convenient for the Bluetooth master to perform signal processing.
Using Metal-Lid LGA package, the overall appearance is a combination of PCB substrate and metal shell. The PCB substrate is located at the bottom, various signal traces are distributed on the front and back of the substrate, and lead-out electrodes are arranged on the back. There are tin-scribing channels distributed around the base plate, so that the top metal shell is connected to the base plate through solder to achieve complete electromagnetic signal shielding. The LGA is mainly composed of MEMS chips and ASIC chips. The MEMS chip is mounted on the PCB substrate, and the ASIC chip is mounted on the MEMS chip to form an upper and lower stack, making the structure more compact and space utilization higher. The signal interconnection between the MEMS chip and the ASIC chip, the ASIC chip and the PCB substrate is realized by means of Wire-Bonding. The package size of 2.7 x 1.8 x 1.1mm is small and compact and compatible with standard microphone sizes.
The device has a sensitivity of -26dBV/g @1kHz, a signal-to-noise ratio of 55dB, and a resonant frequency of 4kHz. Generally speaking, the bandwidth of the speech signal is about 1kHz, so the frequency response of the device is flat within 1.5kHz, which is more convenient for algorithm processing.