Goertek, a major player in acoustics, showcased its latest innovations at its inaugural Innovation Conference held July 17-19 simultaneously at its Weifang headquarters and Qingdao global R&D center. The event, designed to foster internal innovation and cross-departmental collaboration, put on display over a hundred cutting-edge technological advancements in Video, Audio, and AI. Notably, Goertek’s novel acoustic solutions are aiming to solve pain points in the burgeoning AI glasses market, drawing significant attention from industry observers.
As AI glasses gain traction in the consumer electronics market, key challenges related to audio quality and user privacy are coming to the forefront.
Goertek presented an ultralight, thin, and private acoustic solution designed to address this concern. Utilizing a leak-proof design, the solution significantly reduces audio spill, making conversations virtually inaudible to bystanders just 0.5 meters away, protecting the privacy of AI glasses users. The solution incorporates an 18mm*10mm high-performance Rhythm speaker. Its racetrack-shaped diaphragm is engineered to conform to the shape of glasses temples, optimizing space utilization. With enhanced sensitivity and greater volume, coupled with meticulously tuned audio algorithms, this system delivers a more balanced and natural sound experience.
Addressing the challenge of accurate voice capture in complex environments, Goertek introduced a multi-microphone array + VPU (Voice Pickup Unit) sensor solution. This setup dramatically improves noise reduction and speech recognition accuracy. Integrating technologies such as sound source localization, directional audio capture, and multi-microphone intelligent strategies, the solution leverages a state-of-the-art DNN (Deep Neural Network) voice algorithm model. This markedly enhances noise reduction in high-noise environments. Moreover, an edge-based speech separation algorithm improves bi-directional recognition accuracy when both the wearer and another person are speaking simultaneously. Notably, the solution features a custom-developed VPU nose pad module that captures bone conduction voice signals. Combined with bone-air conduction fusion noise reduction algorithms, this allows for further improvements in voice activation and recognition rates in challenging acoustic conditions.
Goertek’s strong R&D capabilities position it to be a key player in acoustic innovation for AI glasses. Goertek boasts a comprehensive suite of edge-based audio algorithm libraries, including ImmeAudio sound enhancement, RVF voice enhancement, GOVI PERCEPTION sound perception, and ACOUSMATIX active acoustics. These algorithms have been deeply optimized for rapid deployment on mainstream chip platforms. To improve the accuracy of solution effectiveness assessments, Goertek has also developed a rapid audio effect validation system. This system enables quick evaluation of product audio performance, allowing for the early identification of design areas that could be improved.
The Goertek Innovation Conference, branded with the “Goerinno” innovation IP, is designed to inspire employee creativity, boost internal motivation, and facilitate technical exchanges across departments. The inaugural conference, themed “Accelerate Innovation,” provided a comprehensive showcase of the company’s strategic direction and the latest technological developments.
Original article, Author: Tobias. If you wish to reprint this article, please indicate the source:https://aicnbc.com/5236.html