Font Size: a A A

Workbench Control System Based On Wireless Speech Recognition

Posted on:2017-10-21Degree:MasterType:Thesis
Country:ChinaCandidate:W W XiaFull Text:PDF
GTID:2358330503968193Subject:Electronics and Communications Engineering
Abstract/Summary:PDF Full Text Request
With the development of speech recognition and wireless communication technology and the improvement of the performance of embedded processor, Wireless voice control has become a trend in the control of operating equipment. Speech recognition technology can be implemented on the embedded system platform and then controls the device with voice,which can not only reduces the frequency of the device keys that people directly operating but also reduce costs. And can be widely used in the field of intelligent home and industrial control, which facilitates the operation of people and improve work efficiency.Firstly, the thesis describes the background and significance of the project, the development and application of speech recognition technology are analyzed, development of wireless communication technology is also analyzed, this thesis introduces the basic structure of the wireless voice control worktable, it describes in detail the basic theory of the speech recognition system, the basic process of speech processing signal in voice recognition, the algorithm of speech parameter, the training algorithm of speech recognition and recognition algorithm, it compares classical DTW algorithm with improved DTW algorithm and uses wavelet to remove noise to improve speech recognition rate, then used MATLAB to carry on the simulation for some algorithm of speech parameter which is adopted in paper.In the aspect of hardware design including STM32 core control module, voice input amplifier module, wireless communication module and stepper motor and linear motor drive module in worktable which complete the hardware design of the whole platform.In the part of software design, it introduces the design the flow chart of the entire software firstly and then the STM32 software environment MDK was briefly introduced.Endpoint detection in speech recognition, feature parameter extraction and optimization and DTW algorithm software design are described mainly, at the same time, introduced the program design of wireless voice communication module and the control worktable module.The whole experiment and analysis are carried out from different distance, quiet environment and noisy environment through the combination of hardware and software. The results show that the design of the wireless voice control platform has achieved good results in the distance of 20 meters and noisy environment in speech recognition. It has referencevalue for speech recognition in the intelligent home and practical engineering applications.
Keywords/Search Tags:Speech recognition, Wireless transmission, Wavelet transform, Dynamic time warping
PDF Full Text Request
Related items