Design Of Vehicle Speech Recognition System For Unspecified People

With the rapid development of computer communication technology, voice recognition technology has gradually penetrated into the industrial, automotive, home appliances, services,medical and other areas of life.The car is the most important means of transport in people's modern life.In order to improve the safety of vehicle driving performance, a variety of car voice equipment came into being.Heavy truck has a large car body size, driving more blind spots and driving complex road characteristics.The use of speech recognition technology in heavy truck cars is limited now.Therefore, a heavy truck vehicle voice recognition system for non-specific people is designed on the background of heavy truck vehicle environment in this paper. It meets the safety and comfort requirements of heavy truck driving.First of all, combining the research status of speech recognition technology at home and abroad, this paper expounds the background and significance of this research.Then the article introduces the theory and simulation of speech recognition system, including system composition,speech signal analysis and processing, double threshold endpoint detection algorithm, MFCC feature parameter extraction and speech recognition algorithm simulation.A speech recognition system based on DTW speech recognition model is designed.Then, based on the simulation of the algorithm, the hardware design scheme of the system is expounded.Focus on the hardware design of signal processing core unit, voice acquisition module,CPLD control module, display control module and external interface module.The software design flow diagram based on hardware environment is given.It elaborates the program preparation flowchart of voice preprocessing, endpoint detection, feature parameter extraction and DTW speech algorithm recognition.On this basis, the software algorithm based on DSP software development environment is transplanted into the hardware system.Finally, through the construction of CCS3.3, simulator, development board between the experimental platform, speech recognition system test experiment based on the DSP has been completed.In the CCS development environment to complete the corresponding voice signal analysis for voice commands that are often used in heavy truck environments. It admitted a number of male and female voice recognition test and displayed the corresponding recognition results by LED module and LCD module.After several experiments it analyzes the recognition rate of the system.The experimental results show that the heavy truck car voice recognition system for voice signal recognition rate can reach more than 90% of this paper. For the common vocabularies used in heavy truck driving,it can be quickly identified and get a higher recognition accuracy The system has the characteristics of fast response, high precision and strong anti-noise ability. It can realize the system identification requirements of automobile heavy truck and also has a very important practical significance and broad application prospects.