Novatek Adopts CEVA Audio/Voice Technology for Its Smart TV SoCs
CEVA, a licensor of wireless connectivity and smart sensing technologies, announced that Novatek Microelectronics, a chip design company, has licensed and deployed the CEVA-X2 Audio DSP, ClearVox voice front-end software, and WhisPro speech recognition software to incorporate always-on far-field voice wakeup and control functionality in its multi microphone smart TV system-on-chip (SoC) lineup.
"We are excited to expand our partnership with Novatek and bring the power of the voice user interface to their smart TV SoCs," said Moshe Sheier, vice president of marketing at CEVA, in a statement. "Our CEVA-X2 DSP together with our ClearVox and WhisPro voice software packages allow Novatek to add new features that create a unique and tailored multilingual voice experience for their customers."
ClearVox incorporates advanced algorithms that cope with different acoustic scenarios and microphone configurations, including optimized software for speaker direction of arrival, multi-mic beamforming, noise suppression, and acoustic echo cancellation, as well as the related firmware and driver software. CEVA's WhisPro operates with ClearVox and offers a CEVA DSP-based speech recognition solution for always-listening devices such as smart TVs, smart speakers, smartphones, and Bluetooth earbuds to interact with cloud-based voice assistants. This holistic integration of voice pre-processing and neural network algorithms delivers a high recognition rate in noisy environments and in far field use-cases, while operating locally on the edge device. CEVA's ClearVox and WhisPro are optimized for the CEVA-X2 high-performance audio/voice DSP for audio applications in smart home, mobile, and automotive products. The CEVA-X2 DSP uses five-way VLIW micro-architecture and offers parallel processing with dual-scalar compute engines, support for Single Instruction Multiple Data (SIMD), and optional floating point units for high-accuracy algorithms
Fluent.ai's technologies for neural network-based speech-to-intent applications have been optimized for CEVA's low-power audio and sensor hub processors targeting wearables, consumer devices, and IoT.
Latest Dolby MS12 implemented on CEVA-BX2 DSPs, providing audio support for smart TVs, set-top-boxes, and more.
CEVA's WhisPro speech recognition is now available with open-source TensorFlow Lite for Microcontrollers implementing machine learning at the edge.