AONDevices, a renowned provider of super low-power, high-performance on-device AI processors and full stack solutions, has made a groundbreaking announcement with the launch of their latest innovation, AONDenoise™. This compact and efficient edge AI denoising technology is set to transform the way we experience speech enhancement in various applications such as hearing aids, wireless earbuds, smartphones, and wearables.
Traditional methods of AI speech enhancement rely on multiple microphones, beamforming, and large Digital Signal Processing (DSP) algorithms, consuming significant memory and power resources. AONDenoise, with its state-of-the-art AI-based denoiser, addresses these concerns by offering one of the smallest denoising algorithms available in the market. This breakthrough algorithm, with just 50K parameters, is significantly smaller than traditional DSP noise reduction methods and current AI-based denoisers, paving the way for ultra-low power and real-time processing devices.
The capabilities of AONDenoise are far-reaching, enabling customizable noise cancellation in a variety of environments. Whether it’s chatter in a crowded restaurant, wind noise, dogs barking, babies crying, or keyboards clicking, this innovative solution delivers crystal-clear audio by effectively targeting and eliminating unwanted noise. The potential applications are immense:
Hearing Aids: AONDenoise provides customizable noise cancellation, enhancing performance in diverse environments and improving the quality of sound for hearing aid users.
Wireless Earbuds: With AONDenoise, wireless earbuds can achieve optimal audio quality in transparency mode, conversation mode, and voice calls, ensuring a seamless and immersive listening experience.
Smartphones: Voice calls on smartphones, especially in outdoor settings, can benefit from the enhanced voice quality delivered by AONDenoise, enabling clearer communication.
Podcasting: AONDenoise enables crystal-clear audio streaming even in noisy surroundings, revolutionizing the podcasting experience for both creators and listeners.
The power of AONDenoise lies in its utilization of AI and deep learning neural networks, enabling it to distinguish human speech from background noise with remarkable accuracy. AONDevices has developed a dedicated software tool suite to maximize the efficiency of the neural network, allowing for customization and training to handle specific noise environments or a broader range of noises. This flexibility empowers both in-house and customer development, opening up possibilities for tailored noise cancellation solutions.
Dr. Youhong Lu, the newly appointed Technical Director at AONDevices, expressed great enthusiasm for the potential impact of AONDenoise on wearable and mobile devices. Dr. Lu, who brings a wealth of experience from esteemed audio engineering roles at industry leaders such as Conexant, Goodix, Plantronics, Verizon, and Microsoft, praised AONDenoise as a remarkable accomplishment in AI speech enhancement. He emphasized the positive effect it will have on countless lives and expressed excitement for the upcoming technology revolution that will be driven by AI, delivering exceptional user experiences.
Integration of the AONDenoise solution into third-party processors is now available, offering seamless adoption for interested customers. Additionally, AONDevices is gearing up to launch the denoiser hardware IP and the highly anticipated AON2100™, the next-generation chip, providing dedicated hardware solutions in the coming months.
With AONDenoise leading the way forward, AONDevices is poised to reshape the landscape of AI speech enhancement, revolutionizing the quality of audio experiences across various devices and applications. The future holds great promise as this compact, efficient, and powerful technology sets a new standard for edge AI and human-machine interfaces.
