Distributed Speech Recognition In Noisy Environment And Vui Design

Nowadays in China, in the process of aircraft maintenance, the related informationis obtained mainly through the on-spot access to maintenance manual and technicalmanual. However, an aircraft is made up of tens of thousands of spare parts. If thetraditional method is employed, it is more likely to miss some information. Moreover,with a great deal of time consumed, the maintenance efficiency is considerably low.In order to make up for the deficiency of traditional approach, this paper, byadopting wireless and distributed technologies, introduces a VUI interface and initiatesa distributed VUI system applied to aircrafts maintenance. By adopting this approach,much processing work is left to background servers. In this way, the working efficiencyof the whole system can be greatly improved and the power consumption of thehandheld terminal is significantly reduced. Meanwhile, in view of the much noise at themaintenance spot, which will cover the voice of maintenance workers, this paperproposes to use double microphones to reduce noise and adopts RIS as the adaptivefiltering algorithm. This approach wins an advantage over the traditional treatment offrequency domain in that it ensures the accuracy of VUI system by effectively reducingthe interference of noise.The VUI system can provide great convenience for the aircraft maintenance. Onthe one hand, maintenance staff can easily finish their routine work only by using thehandheld terminal, without consulting the paper documents and maintenance manuals.On the other hand, the VUI system provides “hands-free” services for the maintenancestaff, allowing them to finish their maintenance work only by using speech. Therefore,the maintenance efficiency can be significantly improved.