The Research And Design Of Speech Recognition On Vehicular Bluetooth Speech Control System

With the popularization and development of computer applications, speech and computer communication as a way of connecting between humans and computers has become the most convenient and efficient means of communication. Speech recognition is the high technology which can make machine be able to understand the human mind. The integration between speech recognition and vehicular intelligent control can not only meet the increasing for automotive functional requirements year by year, but also can made it in a safe, convenient and efficient way to complete entertainment, leisure, work and other activities. Therefore, the research for Vehicular Speech Control System is very important.This paper uses the chip named BlueCore5-Multimedia empoldered by CSR company.Use the speech recognition based on HMM method and design of the speech recognition process of the vehicular speech control system. Through the driver's simple speech commands, vehicular multimedia MOST network nodes are in the intelligent control. By the experiments,the system is proved successful in the functions of the sub-modules and system, to complete the speech training and the speech recognition process.And the system has the good speech training rate and high speech recognition rate.The first chapter sets out the background and significance of the project,introduces the history and development of speech recognition technology and the research status of the vehicular speech control system. Speech recognition technology originated in the 20th century, 50 years.It is a crucial branch of speech signal processing. This technology crosses to all research areas,and the speech is the main object of it. In recent years, chips made of speech recognition technology realized small,lightweight speech system step by step.More and more small speech control systems are used in vehicular systems to realize intelligent control. Some vehicle manufacturers have already introduced a number of new cars of different speech control system.The second chapter researches the basic principles of speech recognition technology. The recognition process includes preemphasis, adding windows ,endpoint detection, feature extraction, template matching. Focuses on two feature extractions,LPC and MFCC. This part analysis of the current three major speech template matching methods, including DTW,HMM and ANN,compares the advantages,disadvantages and application areas of the three models.The third chapter lucubrates the speech recognition technology based on DHMM in this system. Designs every stage algorithm including preemphasis, endpoint detection, feature extraction, template matching based on the actual functional requirements of this system. Discusses parameters in preemphasis and endpoint detection,sets out LPCC feature extraction and DHMM template matching method in this system and describs the basic structure and the specific implementation process. In addition, describes design of the speech command and the production process of the prompt speech.The fourth chapter describes the design process over the vehicular speech control system.This chapter makes a deep study of the performance and features of the master chip BC5-MM and the development environment Bluelab used in software development process.Then the main reason for selecting the chip is set out. Based on the analysis of system requirements and the features of the software system, this part indicates the design and implementation process in the hardware system. At last,the chapter describes the design process of the system, in accordance with the realization of the order,designs training and recognition program in the recogniton part by the realization order.The fifth chapter is part of the system debugging and functional verification, this chapter first sets out the pictures of hardware and development board of the vehicular speech control system.Then three experiments in the speech control process based on the hardware platform are designed.Speech training and speech recognition also the completion of the ovreall system function are tested during these experiments. The results of training rate and recognition rate show that the speech training process and the recognition process are able to complete the job pretty well. Finally, do a series of steps on the system, the results show that the system can work properly, and basically meet the needs of users.The sixth chapter makes systematic summary of the full text, and view the system problems to be studied in the future. Anti-noise optimization and speech matching process optimization can both to be further study. In this paper, the research and design of the speech recognition process based on vehicular Bluetooth speech control system using BC5-MM chip are completed successfully. By experiments it proves that the system has good recognition effect, can meet the needs of control.This paper makes exploration work for the future research and in-depth development of the vehicular control system.