Chinese scientists have achieved a groundbreaking feat in the world of technology by creating what they claim to be the most energy-efficient artificial intelligence (AI) microchips globally. These tiny chips, small enough to fit into devices like smartphones, could pave the way for exciting new features such as offline voice control and mind control.
At the prestigious IEEE International Solid-State Circuits Conference (ISSCC) 2024, Chinese researchers from the University of Electronic Science and Technology of China (UESTC) unveiled their remarkable achievements. This conference, often likened to the “Olympics” of the chip industry, gathers top minds to discuss the latest chip innovations. Held in San Francisco from February 18 to 22, this year’s event saw the debut of these revolutionary AI chips.
One of the chips showcased at ISSCC is designed to enable offline voice control in smart devices. Unlike traditional chips, this one excels in recognizing specific voices even in noisy environments, like when there’s music or conversations in the background. This breakthrough chip consumes very little energy, setting new standards for efficiency and accuracy in voice recognition technology.
ALSO READ: How Chatgpt control robots Using Artificial Intelligence
“The chip achieves a recognition energy consumption of less than two microjoules per instance, with an accuracy rate exceeding 95 per cent in quiet scenes and 90 per cent in noisy environments, setting new global benchmarks for both energy efficiency and accuracy,” the researchers explained.
The second chip presented by the UESTC team is aimed at detecting seizures in people with epilepsy. By using EEG signals, this chip can alert patients in real time, allowing them to seek medical help promptly. Notably, it’s incredibly energy-efficient, making it ideal for use in wearable devices.
“Existing designs rely on extensive patient seizure data for training to achieve high accuracy, a process that is time-consuming and costly due to the low occurrence of seizures and the need for hospitalization,” explained the researchers.
Beyond its primary function, the seizure detection chip has a range of potential applications, including brain-computer interfaces and sleep monitoring. With further advancements, these chips could revolutionize how we interact with technology and monitor our health.
ALSO READ: Artificial Intelligence Can Now Read Your Minds
Before use, patients simply need to wear the device for a two-minute calibration in their natural state, enabling the device to recognise individual signal characteristics.
“This chip also has potential applications beyond seizure detection, including other brain-computer interfaces and sleep monitoring,” the report added.
During a demonstration at the ISSCC, the South China Morning Post (SCMP) reports, EEG signals collected from a wearable brain-computer interface device were transmitted in real-time to a test board using Bluetooth technology.
The chip was reprogrammed to recognize imagined motor commands, which enabled the user to control a robot’s movement. The robot could be commanded to move forward, stop, or move in reverse.
This achievement by Professor Zhou Jun and his team marks a significant milestone in AI chip development. Their innovative approach not only reduces power consumption but also opens doors to new possibilities for smart devices and healthcare technology. The team also has partnerships with leading companies such as SenseTime, Huawei Technologies and electronic components producer BOE.
