CEVA audio/voice DSP and software enable customizable voice wake-up and control capabilities

CEVA, Inc. (NASDAQ: CEVA), the leading licensor of wireless connectivity and smart sensing technologies, announced that Novatek Microelectronics, a leading fabless chip design company specializing in the design, development and sale of a wide range of display driver IC and SoC solutions, has licensed the CEVA-X2 audio DSP, ClearVox™ voice front-end software and WhisPro™ speech recognition software to its family of multi-microphone Smart TV SoCs for always-on, far-field voice wake-up and control.

"We are pleased to expand our partnership with Novatek to bring powerful voice user interfaces to their Smart TV SoCs," said Moshe Sheier, vice president of marketing at CEVA. "Our CEVA-X2 DSP complements the ClearVox and WhisPro voice software packages, enabling Novatek to add new capabilities to create unique, tailored, multi-language voice experiences for its customers."

CEVA's in-house developed ClearVox leverages its extensive expertise in audio and voice processing, combining advanced algorithms to handle different acoustic scenarios and microphone configurations, including optimized software for speaker direction of arrival, multi-microphone beamforming, noise suppression and acoustic echo cancellation, as well as related firmware and driver software. Working in tandem with ClearVox, CEVA's WhisPro provides customers such as Novatek with a powerful speech recognition solution based on CEVA DSPs, enabling always-listening devices such as smart TVs, smart speakers, smartphones and Bluetooth earbuds to interact with cloud-based voice assistant services. This approach is a holistic integration of speech pre-processing and neural network algorithms that can run locally on edge devices, providing high recognition rates even in noisy environments and far-field use cases, thereby protecting user privacy and providing minimal latency. CEVA's ClearVox and WhisPro are optimized for the CEVA-X2 high-performance audio/voice DSP, which is designed for intensive audio applications in the smart home, mobile and automotive markets. The CEVA-X2 DSP uses a 5-way VLIW microarchitecture and provides parallel processing through dual scalar compute engines, support for Single Instruction Multiple Data (SIMD), and an optional floating point unit for high-precision algorithms.