Elderly population (over the age of 60) is predicted to be 1.2 billion by 2025. Most of the elderly people would like to stay alone in their own house due to the high eldercare cost and privacy invasion. Unobtrusive activity recognition is the most preferred solution for monitoring daily activities of the elderly people living alone rather than the camera and wearable devices based systems. Thus, we propose an unobtrusive activity recognition classifier using Deep Convolutional Neural Network (DCNN) and anonymous binary sensors (PIR sensors and door sensors). The real world long-term fully annotated Aruba open dataset collected by binary sensors has been employed for the training and evaluation of the classifier. First, ten basic daily activities, namely: Bed_to_Toilet, Eating, Meal_Preparation, Relax, Sleeping, Work, Housekeeping, Wash_Dishes, Enter_Home, and Leave_Home are segmented with different sliding window sizes, and then converted into binary activity images. Next, the activity images are employed for training the DCNN classifiers with different parameters. Finally, the trained classifiers are evaluated with the 10-fold cross validation method, and the results show that the best DCNN classifier gives 0.79 and 0.951 of F1-score for all 10 activities and eight activities (excluding Leave_Home and Wash_Dishes), respectively.

In distributed speech recognition (DSR), data packets may be lost over error prone channels. A commonly used approach to rectify this is to reconstruct a full frame rate data sequence for recognition using linear interpolation. In this study, an error-concealment decoding method that dynamically adapts the transition probabilities of hidden Markov models to match the frame loss observation sequence is proposed. Experimental results show that a DSR system using the proposed method can achieve the same level of accuracy as a data reconstruction method, is more robust against heavy frame loss, and significantly reduces the computation time.

In distributed speech recognition applications, the front-end device that stands for any handheld electronic device like smartphones and personal digital assistants (PDAs) captures the speech signal, extracts the speech features, and then sends the speech-feature vector sequence to the back-end server for decoding. Since the front-end mobile device has limited computation capacity, battery power and bandwidth, there exists a feasible strategy of reducing the frame rate of the speech-feature vector sequence to alleviate the drawback. Previously, we proposed a method for adjusting the transition probabilities of the hidden Markov model to enable it to address the degradation of recognition accuracy caused by the frame-rate mismatch between the input and the original model. The previous model adaptation method is referred to as the adapting-then-connecting approach that adapts each model individually and then connects the adapted models to form a word network for speech recognition. We have found that this model adaption approach introduces transitions that skip too many states and increase the number of insertion errors. In this study, we propose an improved model adaptation approach denoted as the connecting-then-adapting approach that first connects the individual models to form a word network and then adapts the connected network for speech recognition. This new approach calculates the transition matrix of a connected model, adapts the transition matrix of the connected model according to the frame rate, and then creates a transition arc for each transition probability. The new approach can better align the speech feature sequence with the states in the word network and therefore reduce the number of insertion errors. We conducted experiments to investigate the effectiveness of our new approach and analyzed the results with respect to insertion, deletion, and substitution errors. The experimental results indicate that the proposed new method obtains a better recognition rate than the old method.

Former studies of coding methods based on speech classification usually utilize acoustic knowledge and techniques of speech recognition, or use a nonparametric approach without any apparent acoustic meaning in the classification. The authors propose a simple but effective method for speech classification which is based not only on the acoustic knowledge but also on the coding mechanism itself. A 3.6-kbit/s CELP (code-excited linear prediction) coder based on the classification method is designed. It can be shown that the speech classification coincides with the acoustic phenomena and is effective for CELP coding with adaptive codebook search. Simulation results show that the quality gets better as the computational complexity is reduced, and the pitch contour of the synthesized speech is smoother than that produced by conventional CELP coders.

The robust H-infinity controller design is investigated in this paper for a class of nonlinear singular systems with state and transmission delays. The proposed controller is configured in multiple feedback loops. Based on the interval type-2 T-S fuzzy modelling and parallel distributed compensation techniques, a fuzzy multiple feedback-loop controller with memoryless state and delayed states is proposed to stabilize the system with required H-infinity performance by using the Lyapunov stability theory. The controller gain matrices are able to be calculated by solving the linear matrix inequalities. To check the feasibility of the proposed method, a numerical example is given and the simulation results indicate that the proposed multiple-feedback-loop controller is effective.

This study aims to develop an elderly care system for improving the interpersonal relationship of the elderly with mild cognitive impairment (MCI) by employing the speaker recognition technique and association functionality of social network platforms. Firstly, the speaker recognition units based on the Gaussian Mixture Model (GMM) and Gaussian Mixture Model-Universal Background Model (GMM-UBM) are implemented to identify the visitor via individual input utterance. After the visitor is identified, the proposed system will be linked to the private database and social network platforms to extract the associated message of two parties. Experimental results indicate that the speaker recognition unit based on GMM-UBM achieves the best performance. Finally, five elderly persons are invited to measure the usability of the proposed system. A questionnaire is used to survey the five elderly persons, and the result indicates that the proposed system is highly potentially applicable in improving the interpersonal relationship of the elderly with MCI.