Method and device for inputting computer instruction by virtue of Chinese digital voice

Abstract

The invention discloses a method and a device for inputting a computer instruction by virtue of a Chinese digital voice. The method and the device are mainly used for accurate voice control over various industrial equipment, traffic equipment and military equipment, and are also used for voice control over various smart phones, table computers, household electrical appliances and wearable equipment. The core technology is that a Chinese voice instruction difficult for a computer to accurately identify is converted into a digital voice instruction which can be accurately identified by the computer by virtue of a Chinese character spelling technology. The method and the device have the greatest advantages that a set of unified, normative, complete, efficient and easy-to-use Chinese digital control language is provided, and the strict requirements of the industrial equipment, the traffic equipment and the military equipment on the accuracy, the real-time performance, the briefness, the noise resistance and the faroutness of a voice control instruction are met.

Description

1 technical field [0001] The invention discloses a method and device for inputting computer instructions by using Chinese digital voice, which is mainly used for precise voice control of industrial equipment, transportation equipment, and military equipment (hereinafter collectively referred to as important equipment), and is also used for various smart phones, tablet computers, Voice control of household appliances and wearable devices belongs to the field of information and communication technology. 2 background technology [0002] Precise voice control of important equipment needs to meet the following requirements: [0003] First, accuracy. The recognition accuracy of voice control instructions of important equipment must reach 100%, and no misoperation is allowed. [0004] Second, real-time. Voice control commands for important devices must be issued in real time without any delay. [0005] Third, brevity. Voice control instructions for important devices must be sh...

AI Technical Summary

Problems solved by technology

[0007] Fifth, extreme

[0010] Second, Chinese voice control requires complex steps and operations such as selection, determination, and repetition, and the start and end signals of voice commands must be issued manually, which are difficult to meet real-time requirements

[0011] Third, there are many homophones in Chinese, and the recognition rate is low. Although communicating with important equipment through a sentence can greatly improve the recognition rate, it does not meet the brevity requirement

[0012] Fourth, the recognition rate of each syllable in Chinese is not exactly the same. Some monosyllables (such as "a", "yi", and "wu") have a very low recognition rate in a noisy environment, which does not meet the noise immunity requirements

[0013] Fifth, at present, the online Chinese speech recognition rate has reached a very high accuracy, but in the stand-alone interactive scenario, it has not yet reached a similar accuracy and does not meet the extreme requirements

[0015] First, the number of voices must be minimized

The greater the number of voices, the greater the difficulty of recognition and the lower the recognition accuracy, and vice versa

[0016] Second, the degree of discrimination between individual voices must be increased

[0018] Fourth, homophonic voice commands must be completely eliminated, that is, the pronunciation of two voice commands cannot be the same or similar, otherwise it is easy to cause misoperation

Images