Search

Both automatic speaker identification (ASI) and speech recognition can be utlized now for the control of modern robots. An ASI algorithm can be implemented at a speech interface of the robot to determine the identity of the person allowed to deal with the robot, while speech recognition can be implemented for the interpretation of the order given to the robot. Robustness of the ASI system is a challenging task in the presence of speech degradations such as noise and interference. This study presents a new approach to improve the accuracy of speaker identification in the presence of interference for robot control applications with a convolutional neural network (CNN). First, the speech signal from the speaker is divided into segments, each of which is transformed into a spectrogram, and hence Radon transformation is estimated for this spectrogram. The spectrogram resolves the speech segment into a map of power distribution with both time and frequency. Together, the spectrograms and their Radon transforms are used as inputs to a proposed CNN-based deep learning model. Necessary refinements are undertaken and the resulting optimized “Radon-Deep-Learning Model (RDLM) is compared with a benchmark model. The proposed model consists of six convolutional (CNV) layers followed by six Max. pooling layers, while the benchmark model consists of three CNV layers followed by three Max. pooling layers. Experimental results reveal that the proposed RDLM model achieves a high classification accuracy up to 97.5%, which is more than double the performance reported for some traditional methods that are used for speaker identification.

Epilepsy seizure prediction has become one of the interesting fields that attract researchers to innovate solutions. For epilepsy patients, Electroencephalography (EEG) signals consist of three activities: normal, pre-ictal and ictal. In order to design a prediction model for the ictal state, it is required to distinguish between the activities of EEG signals. This paper presents efficient seizure prediction approaches from EEG signals based on statistical analysis, digital band-limiting filters and artificial intelligence. Band-limiting filters are used to remove out-of-band noise and spurious effects. Then, statistical analysis is adopted for channel selection and seizure prediction based on a thresholding strategy. This statistical analysis depends on amplitude, median, mean, variance and derivative of the EEG signal. The adopted band-limiting filter affects the seizure prediction metrics such as accuracy, prediction time and false alarm rate. The prediction process consists of two phases: training and testing. Both k-means clustering and Multi-Layer Perceptron (MLP) networks are considered for seizure prediction based on artificial intelligence. The proposed approaches can be implemented in a mobile application to give alerts to patients or care givers. The simulation results reveal that the proposed approaches present high performance in terms of accuracy, prediction time and false alarm rate.