Intelligent control method and device

Abstract

The invention provides an intelligent control method and device. The method comprises the following steps that a local storage node acquires the sample data of the target user from acquisition equipment; a computing node or an acceleration node trains the sample data so as to generate a training model; and the computing node controls operation of target equipment according to the training model. The sample data of the target user are stored in the local storage node so that the security of the user data can be further guaranteed; besides, the sample data are trained through the computing nodeor the acceleration node so that the processing capability for the user data can be enhanced, voice and image recognition training is performed on the customized data of the target user, the initial general neural network model can be improved to the training model having extremely high recognition rate for the personal voice characteristics, and thus the voice recognition accuracy and the intelligent degree of the AI platform can be greatly enhanced.

Description

technical field [0001] The invention belongs to the technical field of artificial intelligence, and in particular relates to an intelligent control method and device. Background technique [0002] With the rapid rise of artificial intelligence (AI) technology, the combination of intelligent platforms and the Internet of Things has attracted people's attention. One of the most concerned functions in the field of artificial intelligence is natural language recognition: natural language recognition software has made great progress, such as Siri developed by Apple, Cortana developed by Microsoft, etc. have all appeared on people's smartphones . These softwares are currently positioned as personal intelligent assistants, which can help people control existing functions on mobile phones only through voice input, such as writing memos, arranging schedules, and so on. [0003] However, the current rapid development of Internet of Things technology has enabled many companies to see...

AI Technical Summary

Problems solved by technology

[0003] However, the current rapid development of Internet of Things technology has made many companies see more possibilities, that is, the development of smart homes: smart TVs, network boxes, air conditioners, air purifiers, etc., each device can be equipped with a wifi module or a Bluetooth module. The mobile phone or the remote control communicates, but with the diversification of device functions, there are more and more buttons on the mobile app or the remote control, and the control becomes more and more complicated and cumbersome. When there are more smart devices in the home, in order to realize One job, maybe 2 or 3 remotes to switch back and forth to control

[0004] With the advent of the smart home system, it is possible to control various smart devices (such as smart TVs and smart speakers) in the home through natural language, for example: by "turning on the TV", "changing channels", "off "Light" and other verbal instructions to conveniently control the home devices, but in this solution, the control center is generally served by smart TVs and smart speakers, and such control centers can only implement simple functions, and their computing power is not enough to meet the needs of the users. Existing artificial intelligence systems to train and improve their intelligence

[0005] In order to improve computing power, the cloud obtains various user sample data (such as audio data, video data), trains the AI ​​​​neural network for processing natural language, and then performs natural language recognition through the training model of the cloud platform, but the user Uploading various sample data to the cloud is likely to cause user privacy leakage, making user information security impossible to guarantee

Images