Introduction To The Existing Calculation Of Mass Production!This Manufacturer Is About To

From the official shipment in May 2022, the customer has about 10 products based on the WTM2101 chip. The cooperative customers are mainly intelligent wearable device manufacturers who have higher requirements for computing power, power consumption, and application.EssenceFor nearly one year in the market, Zhicun Technology has shipped nearly one million in customer products such as TWS headsets, VR headsets, and smart watches. 

As a new type of architecture computing power, it can effectively solve the problem of storage walls and power -consuming walls under the traditional von Norman architecture, which is expected to become one of the advanced productivity in the digital economy era.

Recently, the reporter of Xinzhiya entered the industrial chain column interviewed Wang Shaodi, the founder and CEO of Zhicun Technology, the leading enterprise of domestic deposits.In the context of the rapid development of AI chips in Houmore, the current development of technology companies and outlook on the future development of the industry.

The leaders of the integrated industrialization of the deposit, the shipment of shipments in less than a year is close to one million pieces

It is understood that Zhicun Technology was established in 2017. It focuses on the industry's integrated chip field, and is the pioneer and leader of the integrated industrialization of calculations.

As early as 2012, the founding team of the company began to explore Nor Flash -based co -calculation chip technology.After 4 years of research in the laboratory, the company CTO Guo Xinjie has gone through 6 streams and captured the technology of deep learning chip technology for the first time, and made the first internationally of the first floating grid depository deep learning chip.It proves that the existence integration technology can be used in deep learning.

In 2017, the existence integration technology began to rise in the industry, and large international semiconductor design, equipment and production companies all started investing in this technical route.Wang Shaodi realized that the integrated technology of the deposit and calculation was about to open the era of its belonging to it, and then ended the postdoctoral research work in advance to return to China to realize the commercialization of the existence technology.

After more than 2 years of research and development, Zhicun Technology, based on NOR FLASH's integrated chip technology, launched the company's first smart voice chip WTM1001 in 2020.

WTM1001 is the first international calculation -in -the -in -be -in -structure chip product internationally.The power consumption of this chip is 300 μA, and the computing power is 20 times that of similar chips on the market. It can directly store and run neural networks. The calculation process is countless data. The efficiency and the traditional architecture of artificial intelligence chips have been increased by nearly 50 times compared to the artificial intelligence chip.Widely used in low -power AIOT scenarios such as wearable devices and smart terminal devices.

Wang Shaodi introduced.

The first generation of products allowed Zhicun Technology to run through the mass production process of calculating the chip in the deposit, and achieved a breakthrough from academic research to commercial applications.

With the experience of the first generation of products, the company's team continued to start the research and development of the second -generation products.In January 2022, Zhicun Technology announced that the WTM2101 WTM2101 was officially mass -produced and successfully launched to the market in March.

As the world's first SOC chip, WTM2101 is based on the integrated existence technology and uses a small packaging size, which can achieve hundreds of voice command words under the micro -level level.Algorithms, health monitoring and action recognition algorithms, in typical application scenarios, the working power consumption is at the micro -tile level.In addition, the chip can use the SUB-MW-level power consumption to complete the large-scale deep learning operation, which can cover multiple application scenarios in voice recognition, voice enhancement, health monitoring, environmental identification, distant wake-up, sports recognition, visual recognition and event detection.