Applied Brain Research Unleashes Game-Changing Speech AI SDK for Low-Power Edge Devices
Applied Brain Research (ABR) has recently announced the launch of its Speech AI Software Development Kit (SDK) that is now available on the Zephyr Real-Time Operating System (RTOS). This innovative technology is tailored for the Alif Semiconductor's Ensemble family of edge AI processors, marking a significant step forward in the integration of artificial intelligence in low-power, portable devices.
With the new SDK, ABR is set to transform how devices handle voice inputs, significantly enhancing their capabilities without necessitating a cloud connection. This self-contained solution enables on-device streaming speech recognition and text-to-speech features, allowing continuous speech transcription in real-time. The combination of cutting-edge technologies—namely the Niagara Automatic Speech Recognition (ASR) and Nith Text-to-Speech (TTS)—provides a seamless experience with minimal latency and high accessibility.
The newly launched SDK is designed with battery-powered and air-gapped devices in mind, which often face limitations with traditional cloud-reliant voice processing methods. By employing state-space model technology within microcontroller-class silicon constraints, ABR's system achieves superior performance. For instance, on the Ensemble E8, the latency from the first audio input to the first text output is now only 133 milliseconds while consuming a mere 15 milliwatts of additional power. This is a pivotal accomplishment for devices where battery life is critical, enabling always-on voice interfaces that enhance user experience.
Kevin Conley, the CEO of Applied Brain Research, commented, "The Alif Ensemble family combines the low-power requirements necessary for edge devices with the high performance needed to manage real-time speech tasks. The stunning efficiency and real-time capabilities that our SDK offers represent a milestone in what developers can achieve using microcontroller-class platforms."
Mark Rootz, Vice President of Marketing at Alif Semiconductor, also highlighted the increasing preference for the Zephyr platform among developers. He noted, "Voice functionality remains one of the most requested features for battery-operated gadgets. The ABR SDK offers Ensemble developers a practical solution for integrating on-device speech recognition and text-to-speech capabilities, and we are excited to demonstrate this at electronica."
The ABR SDK for Zephyr RTOS is currently available. Developers interested in exploring its capabilities can register for an account and obtain a free evaluation license through the official website. Notably, a live demonstration of the Niagara ASR will take place on the Ensemble E8 at the Alif Semiconductor booth during the electronica 2026 event, running from November 10 to 13 at Messe München.
Founded with the mission of making advanced AI technology accessible in embedded devices, Applied Brain Research is known for its proprietary state-space models that facilitate speech recognition, text-to-speech services, and natural language processing within stringent power and memory limits. For further details, visit www.appliedbrainresearch.com.
Alif Semiconductor, on the other hand, positions itself as a leading supplier of next-generation 32-bit microcontrollers, which blend scalability with performance and power efficiency. Its Ensemble product line redefines the potential of edge computing by merging on-device AI acceleration with robust security and flexible operating systems within a compact design. This unique integration empowers developers to innovate smarter, more efficient devices across various sectors, including consumer electronics and industrial applications.
With the new SDK, ABR is set to transform how devices handle voice inputs, significantly enhancing their capabilities without necessitating a cloud connection. This self-contained solution enables on-device streaming speech recognition and text-to-speech features, allowing continuous speech transcription in real-time. The combination of cutting-edge technologies—namely the Niagara Automatic Speech Recognition (ASR) and Nith Text-to-Speech (TTS)—provides a seamless experience with minimal latency and high accessibility.
The newly launched SDK is designed with battery-powered and air-gapped devices in mind, which often face limitations with traditional cloud-reliant voice processing methods. By employing state-space model technology within microcontroller-class silicon constraints, ABR's system achieves superior performance. For instance, on the Ensemble E8, the latency from the first audio input to the first text output is now only 133 milliseconds while consuming a mere 15 milliwatts of additional power. This is a pivotal accomplishment for devices where battery life is critical, enabling always-on voice interfaces that enhance user experience.
Kevin Conley, the CEO of Applied Brain Research, commented, "The Alif Ensemble family combines the low-power requirements necessary for edge devices with the high performance needed to manage real-time speech tasks. The stunning efficiency and real-time capabilities that our SDK offers represent a milestone in what developers can achieve using microcontroller-class platforms."
Mark Rootz, Vice President of Marketing at Alif Semiconductor, also highlighted the increasing preference for the Zephyr platform among developers. He noted, "Voice functionality remains one of the most requested features for battery-operated gadgets. The ABR SDK offers Ensemble developers a practical solution for integrating on-device speech recognition and text-to-speech capabilities, and we are excited to demonstrate this at electronica."
The ABR SDK for Zephyr RTOS is currently available. Developers interested in exploring its capabilities can register for an account and obtain a free evaluation license through the official website. Notably, a live demonstration of the Niagara ASR will take place on the Ensemble E8 at the Alif Semiconductor booth during the electronica 2026 event, running from November 10 to 13 at Messe München.
Founded with the mission of making advanced AI technology accessible in embedded devices, Applied Brain Research is known for its proprietary state-space models that facilitate speech recognition, text-to-speech services, and natural language processing within stringent power and memory limits. For further details, visit www.appliedbrainresearch.com.
Alif Semiconductor, on the other hand, positions itself as a leading supplier of next-generation 32-bit microcontrollers, which blend scalability with performance and power efficiency. Its Ensemble product line redefines the potential of edge computing by merging on-device AI acceleration with robust security and flexible operating systems within a compact design. This unique integration empowers developers to innovate smarter, more efficient devices across various sectors, including consumer electronics and industrial applications.
Topics Consumer Technology）
【About Using Articles】
You can freely use the title and article content by linking to the page where the article is posted.
※ Images cannot be used.
【About Links】
Links are free to use.