SonicEdge: A New Generation of MEMS Speakers

The SonicEdge speakers provide enhanced audio at low frequencies and extended operation in full audio range and beyond. They are smaller, vibrationless, have a reduced back cavity, are easy to use and integrate, and can be manufactured with high uniformity, similar to semiconductors. SonicEdge’s initial focus is on the TWS and hearable markets, where their speakers enable smaller, more comfortable, and better sounding earphones.

The SonicEdge MEMS speaker is expected to revolutionize the audio industry similarly to how LED technology transformed the lighting industry, MEMS microphones changed the microphone industry, and digital transistors revolutionized electronics in the 60s. The SonicEdge MEMS speaker has several advantages over existing technologies, such as:

  • Miniaturization: The first SonicEdge speaker (SE1000) has a footprint of only 4.0×6.5 mm, which can easily fit into the ear canal and provide freedom to the earphone designer.
  • Scalability: The SE1000 MEMS speaker consists of about 400 identical cells (which will be described later in this articles), this allows the freedom of increasing the sound pressure level (SPL), by increasing the number of cells, and also vise versa, when it is needed to decrease the SPL and for that reducing the number of cells. Additionally, as shape of the MEMS speaker can be changed, the same SPL will be maintained as long as the number of cells is identical.
  • Wide Frequency Response: The SonicEdge speaker covers a wide range of audio frequencies, from low to high. The SE1000 covers the full range of frequencies and enables enhanced ANC performance as it provides extra SPL in the lower frequencies.
  • Durability and Reliability: The SonicEdge speaker is made of robust and durable materials that can withstand mechanical stress, temperature variations, and other environmental factors. This makes it suitable for various applications, including those in harsh environments.
  • Integration with Other Components: The SonicEdge speaker can co-exist with other sensors, such as microphones, accelerometers, or gyroscopes, on a shared PCB since it has minimal mechanical vibrations. This opens up possibilities for advanced audio systems, sensor arrays, and MEMS-based smart devices.
  • Chipset: The SE1000 is a chipset of two components: the MEMS chip and the electronic driver chip. Having two individual chips allows design freedom of the end-application. For example, the MEMS speaker can be in the ear canal without considering the electronic driver chip location.
  • Purely Digital Solution: The audio interface is purely digital, accepting an I2S or PDM signal.

One of the unique features of the SonicEdge MEMS speaker is ultrasound active modulation. The speaker consists of two active membranes: one generating a pure ultrasound signal (the carrier frequency), and another membrane that modulates sound frequency on the same carrier frequency. By mixing these two signals, the speaker produces audible sound frequencies. For example, to generate a 1kHz tone with a carrier frequency of 400kHz, the first membrane would generate a 400kHz frequency, and the second membrane would generate 401kHz (the modulated frequency of 400kHz + 1kHz). The resulting sound would be the 1kHz signal actively demodulated by this mixer. This principle creates a constant volume velocity source or a pump, which explains the pressure build-up at lower frequencies.

The speaker resonance frequency is at the ultrasound regime, which makes it simpler to design acoustic structures to manipulate the SPL response. For example, a cavity with a tube in front of the speaker will introduce a resonance to enhance a range of desired frequencies (say 3-5kHz).

Since the speaker operation principle is a pump, it is ignoring the non-linearity of traditional speakers for different dBFS (dB Full-Scale), which means the spectrum response of the speaker is linear from 0dBFS to -70dBFS. This, together with the fact that the speaker resonance is not in the audio range, allows an application sound engineer to design a simple (linear) equalizer.

Another significant advantage of the SonicEdge MEMS speaker is its minimal phase delay (or group delay) compared to existing technologies like voice coils. This advantage refers to the speaker’s ability to maintain precise phase alignment between the input audio signal and the generated sound waves. Sound waves radiated from the MEMS speaker can recreate the spatial cues and imaging captured in the original audio signal to enhance the listener’s experience by providing a more realistic and immersive soundstage. By reproducing sound signals accurately, SonicEdge MEMS speakers achieve superior sound quality and enable advanced signal processing techniques such as ANC. Additionally, this minimal phase delay preserves the timing and coherence of the audio signal, allowing the speaker to faithfully reproduce complex audio waveforms.

Other markets: The SonicEdge speakers are also suitable for free field applications, such as xR (AR, VR, etc) glasses, mobile phones, automotive and others. For supporting those markets SonicEdge has plans to introduce suitable products in the future.

These unique features make SonicEdge MEMS speakers an innovative solution in the audio industry that offers superior performance and quality compared to traditional speakers.