Allwinner R329 Smart Speaker Processor Features Arm China’s AIPU (Artificial Intelligence Processing Unit)

Allwinner R329

Allwinner R328 is a dual-core Cortex-A7 processor with 64MB or 128MB built-in RAM designed for low-cost smart speakers that was introduced last year and found into smart speaker sold in mainland China. According to a recent press release (in Chinese only), the company has now released a 64-bit update with Allwinner R329 dual-core Cortex-A53 processor equipped with dual HIFI4 DSP for audio post-processing and pre-processing, as well as Arm China’s AIPU (Artificial Intelligence Processing Unit) delivering up to 0.256 TOPS at very low power. There’s no product page for Allwinner R329 yet, so I extracted some specifications from the press release: CPU – Dual-core Cortex-A53 @ up to 1.5 GHz DSP- Dual-core HIFI4 DSP @ 400 MHz AI Accelerator – Arm China AIPU with 0.256 TOPS Built-in DDR RAM Audio Embedded second-generation VAD hardware 5x audio ADCs 2x audio DACs with 100dB SNR I2S and DMIC controller 5-1-channel and 7.1 channel support Integrated dual LDO The company will also provide …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

KDE Plasma Bigscreen for TVs and TV Boxes Offers a Linux Alternative to Android TV

Plasma Bigscreen

KDE Plasma is a desktop environment initially developed for Linux Desktop PC or SBCs, but that’s also available on Linux phones with Plasma Mobile (previously known as Plasma Active). The developers have now decided to work on a version for the big screens with Plasma Bigscreen suitable for TVs and TV boxes and offering an open-source, Linux-based alternative o Android TV. Plasma Bigscreen is community supported, completely free and open-source, and offers an interface fairly similar to Android TV launcher. It is optimized for IR remote control use including (experimental) CEC support, as well as voice control through MyCroft open-source AI & voice assistant and Bluetooth or RF remotes with a built-in microphone such as WeChip G20 or W2, or a USB microphone. By default, Plasmas Bigscreen connects to Mycroft’s Home server, which uses Google’s STT (Speech to text) sending anonymized utterances to Google, but it’s also possible to switch to self-hosted systems such as Mozilla DeepSpeech, or just de-activate …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

This Non-Invasive AI Temperature Screening System Checks Multiple Persons On-the-Fly

Remote AI Temperature Monitoring System

Yesterday, I had to go to the bank with the branch located in a building where they closed all doors except one in order to make sure all people were wearing masks, check their temperature, and let them wash hands with alcohol hand gel. Those are good preventive measures to slow the spread of SARS-CoV-2 virus, but in case many people visit a location that’s also time-consuming. I’ve just come across a system that should speed up the process, as it can check forehead temperature of people as they walk by, using a camera and artificial intelligence for face detection, and an IR camera/thermometer to report temperature. Remote AI temperature screening system specifications: SoC/Memory/Storage – Don’t know, but I would not be surprised if it was based on RK3399 since Rockchip often promotes AI use cases for this processor. External Storage – MicroSD card, USB drive Video Output- HDMI 2.0 output Connectivity – Ethernet & WiFi with external antenna Camera …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

Sipeed M1n is a $10 M.2 Module based on K210 RISC-V AI Processor

Sipeed M1n with Camera & USB-C adapter

Kendryte K210 is a RISC-V processor with AI accelerator found in boards such as Maixduino, Grove AI HAT, or HuskyLens among others, and enabling low-cost, low power AI applications such as face detection or object recognition. You can now add Kendryte K210 AI accelerator to any board or computer with M.2 socket or [Update: the M.2 connector pinout is non-standard] a USB-C port thanks to Sipeed M1n M.2 module that also comes with an M.2 to USB-C adapter. Sipeed M1n specifications: SoC – Kendryte K210 dual-core 64-bit RISC-V processor @ up to 400MHz with FPU, Neural-network Processing Unit (NPU), audio processor, built-in 6MB SRAM memory for CPU, and 2MB AI SRAM Storage – 128Mbit SPI flash Camera – 24-pin connector for DVP camera (OV0328 camera module provided as part of the kit) Host Interface – M.2 socket with some IOs and JTAG interface, accessible via Maix Nano M.2 to USB-C adapter. Supply voltage – 5.0V±0.2V with at least 300mA Temperature …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

UNISOC Unveils T7520 5G SoC Manufactured with 6nm EUV Process Technology

UNISOC T7520

UNISOC has launched its second-generation 5G mobile SoC with T7520 processor manufactured with a 6nm EUV process technology and equipped with four Arm Cortex-A76 cores, four Arm Cortex-A55 cores, as well as an Arm Mali-G57 GPU, and the company’s Makalu 5G platform. T7520 processor is said to support “coverage enhancement for any application scenario, by allowing carriers to deploy 5G on their existing 4G spectrums” (aka NSA = Non-stand Alone). and is optimized for use on 500 km/h high-speed railway. Highlights of the processor include: Advanced 6nm EUV process technology with extremely high lithographic resolution and a better balance between cost, performance and power consumption. Compared to the previous 7nm process, the new process has an 18% improvement in density of transistors and reduces power consumption by 8%. Lower power consumption made possible by the improved process mentioned above as well as an improved low-power consumption architecture and AI-based power regulation technology. For instance, T7520 can reduce power consumption by …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

e-Con Systems Launches MIPI CSI-2 Cameras for Google Coral Dev Board

3.4MP Camera Google Coral Development Board

Google Coral Development Board Camera Series e-Con Systems has announced a camera series that can be used with the Google Coral Development board. There are two models: e-CAM50 CUCRL is a 5.0 MP MIPI CSI-2 camera and e-CAM30 CUCRL is a 3.4 MP MIPI CSI-2 camera. Both Have Features In Common Both support a variety of resolutions and were developed for AI devices. The cameras are color and fixed focus and can be connected directly to the Coral Development Board through P1 and J15 camera connector.  Background on Google Coral Development Board The Google Coral Development board was reported on extensively since its release in 2019. Two articles for the Coral Development Board and its most recent addons are the Google Coral Development Board announcement and the Coral Development Board’s latest mPCIe and M.2 cards. Jetson Nano and Rock960 SBC’s Cameras e-Con Systems previously made cameras for other SBC’s including NVIDIA Jetson TX1/TX2 cameras and cameras for Rockhip RK3399 based …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

MediaTek Helio P95 Processor Launched with APU 2.0 AI Accelerator, Faster Graphics

MediaTek Helio P95

MediaTek has just unveiled an upgrade to its Helio P90 processor. MediaTek Helio P95 still features the same APU 2.0 engine for AI accelerator but with a 10% performance improvement based on ETH Zurich benchmark, and the PowerVR GM 9446 GPU clock has been increased as well to deliver a 10% graphics performance boost. The company also claims a “60% shorter GPU rendering-to-display pipeline”, but it’s unclear whether the comparison is against the P90 processor or other “typical” Arm processors. MediaTek Helio P95 key features and specifications: CPU – 2x Arm Cortex-A75 processors @ 2.2GHz and 6x Arm Cortex-A55 processors @ 2.0GHz using DynamIQ technology GPU – Imagination PowerVR GM 9446 GPU NPU – MediaTek APU 2.0 fusion AI architecture with 1127 GMACs MediaTek CorePilot for sustainable peak performance, longer battery life, and lower operating temperature System Memory –  Up to 8GB of dual-channel LPDDR4x memory @ 1866 MHz Storage – UFS 2.1 storage Display – Full HD+ (2520×1080) resolution, …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

XMOS launches Xcore.AI, a scalable AI processor for the Edge

XMOS, known for its high-performance voice interfaces, is joining the AIoT bandwagon with the announcement of the Xcore.ai, a flexible and economical processor delivering high-performance AI, DSP, control, and I/O’s in a single device. IoT and AI have been one of the most trending topics and fields in the last decade. Both areas have seen large innovations in between them. Deep neural networks have become better, IoT deployment cost has also been greatly reduced, and most importantly, they both have a significant impact on multiple industries. An interesting trend recently is the emergence of applications merging AI and IoT together to form so-called AIoT applications. IoT will be the digital nervous system, while AI will become the brain that makes all the critical decisions which will control the whole system. AIoT has led to the development and deployment of what we call AI processors or AI modules that can be deployed to the edge for high-performance Edge computing applications. An …

Support CNX Software – Donate via PayPal or become a Patron on Patreon