Research on Human-Computer Interaction Mode of Speech Recognition Based on Environment Elements of Command and Control System

摘要

Speech interaction belongs to the category of human-computer interaction, which is one of the new interactive methods developed rapidly in recent years. Since the 1990s, speech recognition and interactive systems such as IBM ViaVoice, Apple Siri, Apple Watch and Google Glass have appeared successively. Speech recognition is characterized by flexibility, multi-channel coordination, sensitivity and temporality. There are many factors that affect the accuracy of speech interaction recognition. In the command and control system, environmental noise (dB), human noise (dB), speech speed (word/minute) and frequency (word/minute) are the four key factors that affect the accuracy of speech recognition, but few people carry out relevant researches. This experiment mainly studies the influence of the above four factors on speech recognition and sets several experimental levels. The experimental results show that both human noise and environmental noise have significant effects on speech recognition control. Within the range of normal speed, speed has little effect on speech recognition. Under the condition of complex environmental noise, the recognition rate of the high-frequency speech signal is higher than that of the low-frequency speech signal. Human noise has a great influence on text information input. The experimental results can provide data support for the design of the command and control system