Design And Implementation Of Smart Vehicle Voice Interaction System Based On Speech Recognition Technology

Conversation is the most important way for people to connect each other.We always consider talking with others as the most effective way to deliver information.Auto speech recognition technology is playing a more and more significant role in software industry.It can make a smooth communication between machine and human beings via easily-operated and efficient speech recognition engine.With the development of car networking,speech recognition technology has been integrated to the vehicle voice system for giving a better experience of human-computer interaction in order to get a higher safety index of drivers.This thesis introduces the design and implementation of smart vehicle voice interaction system.Main works are listed as followed:Firstly,introduction of basic frame and principle of typical speech recognition system and combination of speech recognition technology in other applications.Besides,the introduction of Nuance DDFW(Dragon Drive Framework),core components and communication protocols in DDFW.Secondly,requirements analysis of smart vehicle voice system,and the overall design of smart vehicle voice system,including voice dialing,voice SMS,voice electrical control,voice navigation,POI voice query,smart parking and rating etc.This thesis also analyses the concept of designing and implementing of each module,with elaborations on design requirements,schemes and specific implementations.Using ASR which stands for automatic speech recognition as recognition engine.HMM hidden Markov model in automatic speech recognition technology,and extended with open dictation technology to make it fully integrated with vehicle system so that we can further improve the transaction processing ability of smart vehicle software.Finally,the application of functional test,automatic recognition rate regression test and performance test of smart vehicle voice system core module is introduced.With the verification,we can confirm that smart vehicle voice system can meet the original design,and it works well.This thesis mainly studies on how to make use of the existing speech recognition technology,from a driver’s perspective,to design and implement a very smart and extensive vehicle voice system,and on proper application of speech recognition module,to provide a software only solution.