VeriSilicon Showcased Its Latest Power-Efficient IP Applications at Embedded World 2024
VeriSilicon (688521.SH) today announced its demonstration at Embedded World 2024 in Nuremberg, Germany, booth numbered Hall 4A-518, where it is showcasing customers’ various leading products leveraging the company’s latest technologies and advanced solutions.
VeriSilicon’s one-stop custom silicon services and semiconductor IP licensing services provide customers with intelligent, secure, and highly adaptable solutions, addressing a wide range of key areas such as Artificial Intelligence (AI) and Machine Learning, Internet of Things (IoT), Consumer Electronics and Smart Devices, Data Center and High-Performance Computing, Smart Healthcare, and Automotive Electronics. Highlighted demonstrations include:
- AI and Machine Learning: A high-performance AI-PC with Blue Ocean’s chiplet acceleration card using GPGPU and NPU IPs powered by VeriSilicon, Google Open Se Cura powered by VeriSilicon, Inuitive vision AI processor featuring VeriSilicon’s dual-channel Image Signal Processor (ISP) IP.
- IoT: VeriSilicon’s LE Audio compliant complete Bluetooth Low Energy IP solution, LE Mesh system solution based on VeriSilicon’s BLE technology.
- Consumer Electronics and Smart Devices: A next-generation 8K TV and a leading Smart Camera integrated with VeriSilicon’s Neural Network Processor (NPU) IP, a next-generation Drone featuring VeriSilicon’s Video Processor Unit (VPU) IP, Smart Watches and Smart AR Glasses based on VeriSilicon’s Graphics Processor Unit (GPU) IP and Display Processor IP, Smart Home devices embedded with VeriSilicon’s IPs.
- Data Center and High-Performance Computing: VeriSilicon’s second-generation data center video transcoding platform solution.
- Smart Healthcare: VeriHealthi health monitoring platform based on VeriSilicon’s ZSP Digital Signal Processor (DSP) and BLE IPs.
- Automotive Electronics: Automotive SoC enabled by VeriSilicon’s ISO 26262 compliant IPs and subsystems.
VeriSilicon’s semiconductor IP licensing business holds the largest market share in China and ranks seventh worldwide1. Among the top ten IP companies globally, VeriSilicon’s number of IP categories ranks top two2. With over two decades of expertise in the GPU sector, VeriSilicon’s GPU IP has been shipped in nearly two billion chips. Additionally, VeriSilicon’s self-developed NPU IP has been integrated into 128 AI SoCs supplied by 72 licensees in more than 10 market segments, and has been shipped in over 100 million AI-enabled chips worldwide. In terms of video processing, VeriSilicon’s first-generation video transcoding acceleration solution, which is integrated with the company’s VPU IP, provides 6 times the transcoding capability of traditional high-end CPUs while consuming only 1/13 of the power. This innovative technology has been successfully applied to media accelerator chips custom-made for leading global chip companies based on the 5nm process and has entered the mass production stage.
As part of the Embedded World conference program, VeriSilicon’s Vice President of Vision Image Product Shang-Hung Lin will deliver a keynote speech themed Implementing Transformer Neural Networks for Visual Perception on Embedded Devices during Session 7.8 on Embedded Vision & Edge AI, scheduled for April 10 at 1:45pm.
“We are thrilled to present our latest and innovative IP solutions at Embedded World 2024. Our comprehensive Glass-to-Glass (from camera-in to display-out) intelligent pixel processing IP portfolio is silicon-proven, production-proven, and application-proven, which is configurable for a wide range of applications, ranging from low-power-focused to high-performance data center uses,” said Wei-Jin Dai, Executive VP and GM of IP Division at VeriSilicon. “In addition to our own IP offerings, we have also collaborated with third-party partners such as RISC-V IP providers and GUI software providers to deliver competitive solutions to our customers. Our commitment to IP innovations is aimed at empowering our customers’ success, enabling them to introduce innovative products to the market more quickly.”