AON Devices develops AI speech enhancement algorithms for wearables
AON Devices, a company specializing in on-device edge AI processors, has developed AON Denoise, an efficient denoising algorithm. The compact solution improves performance by utilizing edge computing techniques while minimizing power consumption.
AON Devices claims that their AI speech enhancement algorithm efficiently meets the unique demands of applications like hearing aids, wireless earbuds and wearables that require low latency.
The company says that the AON Denoise solution allows customers to personalize the AI speech-denoising algorithm by training it to handle different noise environments effectively. This solution utilizes AI and deep learning neural networks to distinguish human speech from background noise. The algorithms can detect noises like chatter in a crowded area, wind noise, dogs barking and babies crying.
“A small yet incredibly effective algorithm that works with a single microphone will positively impact many lives. I have been incredibly impressed by AON Devices’ achievements and am excited for the upcoming technology revolution in which AI delivers better user experiences,” says Dr. Youhong Lu, the technical director at AON Devices.
AON Devices claims that its AI denoising solution has a compact design of 50,000 parameters, which makes it significantly smaller than traditional DSPs. The company says it surpasses current AI-based denoising algorithms by an order of magnitude.
AON Devices claims to address the challenges associated with existing AI denoisers’ memory and power consumption with the proposed AON Denoise algorithms. The company is expanding its product portfolio from edge AI human-machine interfaces to advanced human-human interfaces.
AON Devices announced its full-stack edge AI solution earlier this year. The solution combines hardware and software technologies to help manufacturers reduce development time.
“Thanks to the efficiency of our tools, we can produce models for new use cases in a matter of hours or days instead of weeks or months as typically seen in other solutions,” says Germaine Ewing, the business development director at AON Devices.
Article Topics
AI/ML | algorithm | AON Devices | audio | chip | denoiser | edge AI | wearables
Comments