What used to be a one-size-fits-all approach to noise cancellation headphones is getting replaced with algorithms that understand the sonic characteristics of different environments. Whether a user is in a coffee shop, on a train, or walking through the noisiest environment, next-generation audio devices know how to adapt.
What makes this possible is on-device machine learning (ML) that operates in real time. Rather than relying on static noise profiles, artificial intelligence (AI) can enhance an audio system by using microphones, motion sensors, and clues, adjusting what the user hears and doesn't hear.
The results include clearer calls, sharper music, and nearly natural listening experiences.
In this week’s New Tech Tuesdays, we explore how real-time AI-powered noise cancellation can create an adaptive, customized audio experience.
Engineers are now designing earbuds and headphones that can listen to their surroundings and adapt accordingly to provide the best audio experiences. These intelligent systems are made possible by AI-enhanced components like microelectromechanical systems (MEMS) microphones, digital signal processors (DSPs), and advanced Bluetooth®-enabled system-on-chips (SoCs). For example, MEMS microphones can pick up environmental noise, and DSPs process that input on the device with trained ML, but when paired with Bluetooth SoCs, it creates fast and local audio adjustments with low-latency performance. This approach also preserves battery life.
Traditional active noise cancellation (ANC) uses pre-set filters to cancel constant noise, but these filters don’t adapt in real time. AI is changing this fixed-response noise cancellation.
The latest S5 and S3 Gen 2 Sound Platforms from Qualcomm® are designed to provide smarter and more personalized audio to wireless earbuds and headphones. The platforms include motion sensors, digital microphones, and DSPs that can process live audio. These components work together to determine how a user is wearing the device, sense their movement, and figure out where they are. These smart audio systems can then adjust noise filtering as needed, whether on a plane, in a coffee shop, or taking a walk through the city (Figure 1).[1]
Figure 1: A user on a busy train might experience an elevation in their earbud volume with next-gen sound platforms incorporated into the design. (Source: Halfpoint/stock.adobe.com)
These platforms often include software development kits that allow engineers to adjust ANC models for specific use cases or form factors, so they can change how it behaves in earbuds versus over-ear headphones.
Companies are now integrating AI or personalized listening experiences into their products. For example, Sony’s Adaptive Sound Control tracks user location and activity levels and adjusts ambient noise and audio modes in real time.[2] Additionally, Bose's QuietComfort Ultra lineup uses head-tracking to deliver spatial “Immersive Audio” that moves with you.[3]
Engineers are working behind the scenes to design products with these features, incorporating components that support edge AI processing. Low-power microcontrollers, embedded AI Bluetooth SoCs, and digital MEMS microphones allow for real-time analysis and filtering without cloud assistance, making them ideal for mobile use, where power efficiency and speed are critical. These innovations help create earbuds that do not just block sound but improve what you hear.
While audio is currently reaching a point of personalization, the future will be hyper-personalized. Engineers are designing ways to include biometric data, like heart rate or skin temperature, into the sound profile to assess for stressful environments and adjust for the user’s benefit. Another aspect of this evolving technology is hearing aids. Valencell is a company that has been working on integrating sensor technologies into hearing aids that could combine biometric sensing with hearing enhancement to autonomously detect stress levels and then automatically adjust audio levels tailored to a user's hearing preferences.[4]
As AI is incorporated further into audio systems, expect improved switching between modes, better speech detection in noisy spaces, and AI-powered beamforming for voice assistants. With beamforming, multiple microphones and different algorithms are used to focus on and analyze audio from the speaker’s direction and suppress background noise.
In the long term, AI-enhanced ANC may expand beyond earbuds to applications like automotive cabins, conference rooms, or even smart homes, where intelligent sound management is part of everyday living.
This week’s New Tech Tuesdays showcases the TDK InvenSense T5848 Digital MEMS Microphone. This MEMS microphone features a signal-to-noise ratio (SNR) of 68dBA and a high acoustic overload point (AOP) of 133dB SPL, enabling it to detect sound even in very noisy environments. In addition, the T5848 includes an acoustic activity detect (AAD) feature that allows designers to program the device to listen for specific acoustic activity amid background noise. Lastly, its I²S interface enables designers to integrate intelligent sound detection features into next-generation audio.
Bringing AI into audio systems changes how engineers approach their designs. There’s a heightened focus on component selection in this sector and DSP programming that makes these devices more adaptable and personalized. For users, these advances mean smarter sounds that fit their world, wherever they are.
Sources
[1] https://www.qualcomm.com/news/releases/2022/11/new-qualcomm-s5-and-s3-gen-2-sound-platforms-set-the-benchmark-f [2] https://www.sony.co.uk/electronics/support/articles/00224928 [3] https://www.bose.com/p/headphones/bose-quietcomfort-ultra-headphones/QCUH-HEADPHONEARN.html?srsltid=AfmBOorv3zQXoGxJ3jwFCkcfrFxvdJlA9n7Kwggsdis_Dtq-7e5MdX05 [4] https://valencell.com/news/making-biometrics-universal-in-hearables-and-hearing-health/
Mouser Electronics, founded in 1964, is a globally authorized distributor of semiconductors and electronic components for over 1,200 industry-leading manufacturer brands. This year marks the company's 60th anniversary. We specialize in the rapid introduction of the newest products and technologies targeting the design engineer and buyer communities. Mouser has 28 offices located around the globe. We conduct business in 23 different languages and 34 currencies. Our global distribution center is equipped with state-of-the-art wireless warehouse management systems that enable us to process orders 24/7, and deliver nearly perfect pick-and-ship operations.