Design of the Speech Control System for a Upper Limb Rehabilitation Robot Based on Wavelet De-noising

摘要

In order to help stroke patients to go on rehabilitation training, a new type of exoskeleton upper limb rehabilitation robot based on biomechanics has been designed in this paper, and speech control has been implemented. The speech control is suitable for patients who can speak. Doctors or family members can also assist patients in training. The system can help patients with hemiplegia to recover the upper limbs and fingers movement function, at the same time, it has the characteristics of small size, more portable and convenient operation. It can also effectively solve the shortage of traditional doctors and patients' one to one rehabilitation training. It is convenient for patients to use in the family. Anybody can help patients with rehabilitation training and has a broad application prospect in the field of rehabilitation. Finally, the speech control system was tested and the master-slave control experiment was implemented. In order to better perform speech recognition and further reduce the noise in the environment, the original speech signals have been de-noised with wavelet transform. The experimental results showed that the success rate of post-noise speech control system rehabilitation training is higher, and the average correct recognition rate of speech control is 94%, thus the feasibility of the speech control system is verified.