Ambiq's AI Model Zoo is a library of reference AI models optimized for Apollo4 Plus, including Speech UI, sound classification and processing, activity recognition, and more. With the addition of neural network speech enhancement based on this open-source TinyML AI model, the low power Apollo4 Plus platform is now able to operate noise removal in real-time with minimal latency and energy.
The Neural Network Speech Enhancer (NNSE) solution, created from the ground up, is built on Ambiq's neuralSPOT AI SDK, and includes all the code needed for training, testing, and deploying on the Apollo4 Plus SoC. This enables developers to create AI-based de-noise applications in minutes.
As Ambiq confirms, NNSE consists of a simple graphical user interface allowing users to easily record and save the enhanced speech along with the original noisy audio on their PC for demonstration purposes. NNSE can capture clean speech for various applications, such as voice memo recording, voice chat, and speech recognition.
This AI model is optimized to operate on devices, in real-time, with minimal latency and energy utilization. While the pre-trained model is ready to use on Ambiq development platforms, NNSE also includes software to train, convert, and deploy customized models where needed. All software has been released under the permissive BSD-3-clause license for ease of deployment and development.
"Ambiq's neural network speech enhancer may be the only open-source TinyML implementation of AI-based speech de-noising for IoT endpoint devices," says Carlos Morales, the VP of AI at Ambiq. "The highly optimized AI model will help developers get started on speech de-noising applications on Ambiq Apollo4 Plus SoC in a matter of minutes."
The Austin, Texas low-power semiconductor company believes this solution will represent a multibillion-dollar opportunity, expanding applications where AI is already used extensively for everything from industrial anomaly detection to speech-based consumer interfaces. Ambiq's AI development on the Apollo SOCs breaks free from the limitations of power consumption and form factor of endpoints to leverage intelligence everywhere without physical constraints.
Developers can download the Technical Preview at github.com and start working on new AI projects.
Ambiq is focused on developing the lowest-power semiconductor solutions to enable intelligent devices everywhere, with devices that last weeks on a single charge (rather than days) while delivering a maximum feature set. Ambiq's goal is also to take Artificial Intelligence (AI) to mobile and portable devices, using Ambiq's advanced ultra-low power system on chip (SoC) solutions. Ambiq has shipped more than 200 million units of these low-power units as of March 2023.
www.ambiq.com
