Design Of Intelligent Vehicle Voice Control System Based On ROS

In recent years,intelligent speech technology is one of the fastest technology in artificial intelligence,and it is also focus of research at home and abroad.In this paper,based on the current popular application of intelligent voice technology,with the help of mature and stable new robot operating system(ROS),speech recognition technology is added to the traditional robot control system,and a set of intelligent voice control system solution based on robot operating system is designed.In this paper,the voice control technology has been studied.Through the analysis of the technology needed in the system design,the specific functions of the system have been determined.According to the development trend of intelligent vehicle system and the difficulties of speech recognition,the design requirements and objectives of the system are determined,the design requirements and objectives of the system are clarified.On this basis,the overall design of the system and the selection of the hardware scheme are completed.In view of the pollution of various types of noise to the speech recognition process in the complex industrial life situation,the received speech signal is processed by speech enhancement.Then the algorithm of speech enhancement is compared,the advantages and disadvantages of Wiener filter enhancement and generalized sidelobe canceller enhancement are analyzed,and the speech enhancement method based on the latter is selected.At the same time,a GSC based adjustable delay matrix method is proposed to overcome the disadvantage of traditional generalized sidelobe canceller.According to the simulation results and the actual measurement results of the software,it can be concluded that in the complex scene with different noises,this method can effectively improve the speech recognition rate while maintaining a certain speech quality,and has good performance.Finally,the system software is designed to complete the corresponding functions: according to the software development process,firstly build Ubuntu system on raspberry PI,and then configure the development environment of ROS.After the collected multi-channel speech signal is transmitted to the speech enhancement node for speech enhancement and noise reduction,the speech recognition node developed based on iFLYTEK speech recognition SDK receives speech and recognizes it as text,which is published by topic.The voice control node receives the corresponding voice control instructions by subscribing to this topic to complete the control of the robot.The simulation results show that the improved GSC generalized sidelobe canceller has better filtering effect than the traditional GSC algorithm under the influence of diffuse noise.The results show that the intelligent speech control system can realize speech recognition,enhancement,control and other functions.Compared with the single microphone scheme,this scheme has high speech recognition rate,can greatly suppress the environmental noise,and has more flexible voice control instructions and high intelligence.