Serial structure is applied to speaker recognition to reduce the algorithm delay and computational complexity.The speech is first classified into speaker class,and then searches the most likely one inside the class.Di...Serial structure is applied to speaker recognition to reduce the algorithm delay and computational complexity.The speech is first classified into speaker class,and then searches the most likely one inside the class.Difference between Gaussian Mixture Models(GMMs) is widely applied in speaker classification.The paper proposes a novel mean of pseudo-divergence,the ratio of Inter-Model dispersion to Intra-Model dispersion,to present the difference between GMMs,to perform speaker cluster.Weight,mean and variance,GMM’s components,are involved in the dispersion.Experiments indicate that the measurement can well present the difference of GMMs and has improved performance of speaker clustering.展开更多
Dear Editor,This letter investigates the system development of a multi-joint rehabilitation exoskeleton,and highlights the subject-adaptive control factors for efficient motor learning.In order to enable the natural m...Dear Editor,This letter investigates the system development of a multi-joint rehabilitation exoskeleton,and highlights the subject-adaptive control factors for efficient motor learning.In order to enable the natural mobility of the human upper extremity,we design the shoulder mechanism by arranging three rotational joints with acute angles,and adopt a serial chain structure for the fully constructed system.After the kinematics and dynamics of CASIA-EXO are modelled,the patient-in-the-loop control strategy is proposed for rehabilitation training,consisting of the intention-based trajectory planning and performance-based intervention adaptation.Finally,we conduct experiments to validate the efficacy of the control system,and further demonstrate the potential of CASIA-EXO in neurorehabilitation.Introduction:Neurological diseases are the leading cause of nontraumatic disability worldwide,and stroke is one of the most common encountered neurological injury,which is suffered by over 15 million individuals each year,and about 70%−80%of these individuals have varying degrees of functional impairments[1].In order to facilitate the motor relearning in central nervous system,post-stroke patients need to undergo long-term rehabilitation training to promote neural plasticity,thereby enhancing the recovery of motor function in activities of daily living(ADLs).Evidence in the clinical studies suggests that robot-assisted rehabilitation integrating neuroscience,biomechanics,and automation control can improve the patients’motivation for active participation while improving the treatment efficiency,therefore,be expected to become the most promising means for neurorehabilitation[2].展开更多
文摘Serial structure is applied to speaker recognition to reduce the algorithm delay and computational complexity.The speech is first classified into speaker class,and then searches the most likely one inside the class.Difference between Gaussian Mixture Models(GMMs) is widely applied in speaker classification.The paper proposes a novel mean of pseudo-divergence,the ratio of Inter-Model dispersion to Intra-Model dispersion,to present the difference between GMMs,to perform speaker cluster.Weight,mean and variance,GMM’s components,are involved in the dispersion.Experiments indicate that the measurement can well present the difference of GMMs and has improved performance of speaker clustering.
基金supported in part by the National Key Research and Development Program of China(2022YFC3601200)the National Natural Science Foundation of China(62203441,U21A20479)+1 种基金the Beijing Natural Science Foundation(L232005)the Inner Mongolia Autonomous Region Science and Technology Plan(2023YFDZ0042).
文摘Dear Editor,This letter investigates the system development of a multi-joint rehabilitation exoskeleton,and highlights the subject-adaptive control factors for efficient motor learning.In order to enable the natural mobility of the human upper extremity,we design the shoulder mechanism by arranging three rotational joints with acute angles,and adopt a serial chain structure for the fully constructed system.After the kinematics and dynamics of CASIA-EXO are modelled,the patient-in-the-loop control strategy is proposed for rehabilitation training,consisting of the intention-based trajectory planning and performance-based intervention adaptation.Finally,we conduct experiments to validate the efficacy of the control system,and further demonstrate the potential of CASIA-EXO in neurorehabilitation.Introduction:Neurological diseases are the leading cause of nontraumatic disability worldwide,and stroke is one of the most common encountered neurological injury,which is suffered by over 15 million individuals each year,and about 70%−80%of these individuals have varying degrees of functional impairments[1].In order to facilitate the motor relearning in central nervous system,post-stroke patients need to undergo long-term rehabilitation training to promote neural plasticity,thereby enhancing the recovery of motor function in activities of daily living(ADLs).Evidence in the clinical studies suggests that robot-assisted rehabilitation integrating neuroscience,biomechanics,and automation control can improve the patients’motivation for active participation while improving the treatment efficiency,therefore,be expected to become the most promising means for neurorehabilitation[2].
