Ambiq has introduced neural network speech enhancer (NNSE), its recent addition to its neuralSPOT’s model zoo. This optimised AI (artificial intelligence) model productively removes background noise from speech on device in real time, allowing clean speech capture in noisy environments. As with all Ambiq model zoo components, NNSE includes scripts and tools to help developers add speech de-noising capabilities to their applications. It also consists of a simple graphical user interface allowing users to easily record and save enhanced speech along with original noisy audio on their PC (personal computer) for demonstration purposes.
Speech de-noising is useful in noisy or loud environments, such as vehicle cabins, factory floors, offices, and outdoors. NNSE can capture clean speech for various applications, such as voice memo recording, voice chat, and speech recognition. This AI model is optimised to operate on devices, in real time, with minimal latency and energy utilisation. While pre-trained model is ready to use on Ambiq development platforms, NNSE also includes software to train, convert, and deploy customised models where needed. All software has been released under permissive BSD (berkeley source distribution)-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. The highly optimised AI model will help developers get started on speech de-noising applications on Ambiq Apollo4 Plus SoC in a matter of minutes.” says Carlos Morales, vice president of AI at Ambiq.
AI is used multibillion-dollar IoT (internet of things) space for everything from industrial anomaly detection to speech-based consumer interfaces. Ambiq’s recent AI development on Apollo SOCs (system on chips) broke free the limitations of power consumption and form factor of endpoints to enable an environment where users can leverage intelligence everywhere without physical constraints.
For more information visit here.
Comment on this article below or via Twitter: @IoTNow_OR @jcIoTnow
