With the concept of smart home becoming more and more popular, intelligent audio has become one of the hottest topics in the AI ​​field, and it is a key investment area for domestic technology manufacturers. As one of the important entrances to control smart homes, intelligent audio has natural voice, interaction and home attributes compared to mobile phones, TVs, tablets and other products, which determines its entry point for future smart life.

From the current point of view, whether it is a global technology giant or a domestic manufacturer, in the past one or two years, the “plus yards” intelligent voice interaction system has been successively launched. Strategy Analytics estimates that about 60 million households in the world currently have at least one smart speaker. It is predicted that by 2024 and 2025, the global smart speaker market will reach 30 billion yuan and 40 billion US dollars. As a home device, smart speakers have a broad prospect of "one machine, one household".

Faced with this trend, the highly intelligent future lifestyle of home will greatly change the demand for data storage for smart audio. In view of this broad application prospect, Hongwang Semiconductor ICMAX is a tailor-made storage solution for smart audio. The eMMC products with a new generation of NAND Flash as storage media have superior performance, mainly in the wide range of ICMAX eMMC capacity selection, fast speed, and use. Long life and better meet the needs of intelligent audio storage. LPDDR is an internal memory that exchanges data directly with the CPU. ICMAX LPDDR has faster speed and lower power consumption, allowing the smart speaker to respond faster.

The internal structure of AI intelligent audio is clearer than the internal structure of other intelligent products. The core of the technology is concentrated on one PCB board. The most important components are SOC and embedded storage eMMC and LPDDR, which are in the whole intelligent sound working process. Play an important role.

ICMAX is applied to intelligent audio modules

Let's take a brief look at how smart speakers provide content and services: for example, to say something to a smart speaker, a microphone to record voice, stored in LPDDR, the CPU reads data from LPDDR, and then uploads the processed digital signal to the cloud over the network. The server; the cloud server will perform voice digital code recognition and semantic understanding, and then the target instructions obtained in the cloud will be transmitted to the smart speaker through the network and stored in the LPDDR, and the CPU will read the data from the LPDDR and execute the corresponding instructions.

Throughout the process, it can be clearly seen that LPDDR plays a vital role in the operation of intelligent sound. As the storage carrier of instruction data, the reading speed of LPDDR directly affects the speed of exchanging data with the CPU and affects the use of intelligent audio. Experience. Nowadays, a single intelligent sound can no longer meet the needs of the present. The intelligent sound with the screen will also be one of the development trends of the future sound. To realize the sound, frequency, picture and text, it needs to be applied to the large-capacity storage medium eMMC. For the general intelligent sound SLC Nand-flash can meet the demand.

Hongwang Semiconductor ICMAX has rich industry experience in intelligent audio storage, which can bring customers specialization and cost-effective storage options, which is more conducive to the realization of low cost, small size, convenient movement, low power, etc. Features to better respond to the needs of intelligent homes.