The voice solution provides superior far-field processing and third-party compatibility, enabling rapid deployment of voice interfaces with top-tier audio quality.
XMOS has launched xcore-voice, a comprehensive voice solution based on the xcore platform’s hardware, software, and tools. This solution showcases voice pipeline designs using XMOS’s established audio front-end, integrating far-field processing and compatibility with third-party Independent Software Vendor (ISV) voice algorithms. This adaptable platform functions as a voice processor by offloading the voice processing tasks from the System on a Chip (SoC). This allows product developers to implement voice interfaces rapidly and affordably, ensuring the best audio quality. The solution provides ready-to-use design examples, harnessing the processor’s power via the platform software and forming the foundation for numerous applications.
The press release says that the platform stands out in three significant areas. First, in terms of power, the solution boasts a low-power architecture that significantly reduces power consumption. Introducing a true standby mode provides the ‘always-on’ functionality many modern electronic devices demand, consumes less than 55mW, and complies with EU energy-saving standards. Secondly, in performance, the voice solution, while harnessing the company’s industry-acclaimed high-performance far-field algorithm, offers a dynamic voice processing workload. This package allows for customisation and feature enhancements. Application-centric software elements like tunable algorithms and foundational software components like FreeRTOS are easily accessible to enhance performance. Lastly, the platform is notably flexible. The blend of general-purpose Digital Signal Processing (DSP), Artificial Intelligence (AI), and Input/Output (IO) processing is software-customisable, considerably cutting down the time needed to launch custom silicon.
For designers, the integration processes have been streamlined, yet it still provides a comprehensive, Bill Of Materials (BOM) optimised solution for integration. The platform offers a seamless voice interface with the capability for local command recognition, reducing reliance on cloud-based systems. This enhances the user experience through decreased latency and increased privacy and reduces cost and power requirements in product designs. Such innovations pave the way for a revolutionary voice-based user experience in upcoming ‘smart’ electronic devices.
“As consumers, we’re increasingly acclimated to the use of voice technologies in our everyday lives – and that makes us more aware of their limitations,” commented Aneet Chopra, EVP Marketing and Product Management, XMOS. “To avoid falling behind rival products, product designers need to be able to deliver ‘across-the-room’ voice interfaces that offer the best possible quality without compromising on speed-to-market or cost.
For more information, click here.