Andoni Arruti

Natural Interaction between Avatars and Persons with Alzheimer's Disease

In this paper a natural human computer interaction paradigm is proposed for persons with cognitive impairments such as Alzheimers Disease. The paradigm consists of using a realistic virtual character, rendered on a common television set, to play the role of a virtual personal assistant that shows reminders, notifications and performs short dialogues with the user. In this paradigm, the television remote control is used as a return channel to capture the users responses. To test this concept, a functional prototype was built and then validated by a group of 21 persons with Alzheimers Disease ranging from mild to moderate. For this validation two simple dialogues were developed that consisted of simple Yes/No type questions. The test results showed that with both dialogues all users engaged naturally with the avatar. All of the users understood the information conveyed by the avatar and answered successfully by means of the TV remote control.

Classifier Subset Selection to construct multi-classifiers by means of estimation of distribution algorithms

Abstract This paper proposes a novel approach to select the individual classifiers to take part in a Multiple-Classifier System. Individual classifier selection is a key step in the development of multi-classifiers. Several works have shown the benefits of fusing complementary classifiers. Nevertheless, the selection of the base classifiers to be used is still an open question, and different approaches have been proposed in the literature. This work is based on the selection of the appropriate single classifiers by means of an evolutionary algorithm. Different base classifiers, which have been chosen from different classifier families, are used as candidates in order to obtain variability in the classifications given. Experimental results carried out with 20 databases from the UCI Repository show how adequate the proposed approach is; Stacked Generalization multi-classifier has been selected to perform the experimental comparisons.

Fusing multiple image transformations and a thermal sensor with kinect to improve person detection ability

This paper proposes a novel approach to combine data from multiple low-cost sensors to detect people in a mobile robot. Robust detection of people is a key capability required for robots working in environments with people. Several works have shown the benefits of fusing data from complementary sensors. The Kinect sensor provides a rich data set at a significantly low cost, however, it has some limitations for its use on a mobile platform, mainly that people detection algorithms rely on images captured by a static camera. To cope with these limitations, this work is based on the fusion of Kinect and a thermical sensor (thermopile) mounted on top of a mobile platform. We propose the implementation of an evolutionary selection of sequences of image transformation to detect people through supervised classifiers. Experimental results carried out with a mobile platform in a manufacturing shop floor show that the percentage of wrong classified using only Kinect is drastically reduced with the classification algorithms and with the combination of the three information sources. Extra experiments are presented as well to show the benefits of the image transformation sequence idea here presented.

Classifier Subset Selection for the Stacked Generalization Method Applied to Emotion Recognition in Speech

In this paper, a new supervised classification paradigm, called classifier subset selection for stacked generalization (CSS stacking), is presented to deal with speech emotion recognition. The new approach consists of an improvement of a bi-level multi-classifier system known as stacking generalization by means of an integration of an estimation of distribution algorithm (EDA) in the first layer to select the optimal subset from the standard base classifiers. The good performance of the proposed new paradigm was demonstrated over different configurations and datasets. First, several CSS stacking classifiers were constructed on the RekEmozio dataset, using some specific standard base classifiers and a total of 123 spectral, quality and prosodic features computed using in-house feature extraction algorithms. These initial CSS stacking classifiers were compared to other multi-classifier systems and the employed standard classifiers built on the same set of speech features. Then, new CSS stacking classifiers were built on RekEmozio using a different set of both acoustic parameters (extended version of the Geneva Minimalistic Acoustic Parameter Set (eGeMAPS)) and standard classifiers and employing the best meta-classifier of the initial experiments. The performance of these two CSS stacking classifiers was evaluated and compared. Finally, the new paradigm was tested on the well-known Berlin Emotional Speech database. We compared the performance of single, standard stacking and CSS stacking systems using the same parametrization of the second phase. All of the classifications were performed at the categorical level, including the six primary emotions plus the neutral one.

APyCA: Towards the automatic subtitling of television content in Spanish

Automatic subtitling of television content has become an approachable challenge due to the advancement of the technology involved. In addition, it has also become a priority need for many Spanish TV broadcasters, who will have to broadcast up to 90% of subtitled content by 2013 to comply with recently approved national audiovisual policies. APyCA, the prototype system described in this paper, has been developed in an attempt to automate the process of subtitling television content in Spanish through the application of state-of-the-art speech and language technologies. Voice activity detection, automatic speech recognition and alignment, discourse segment detection and speaker diarization have proved to be useful to generate time-coded colour-assigned draft transcriptions for post-editing. The productive benefit of the followed approach heavily depends on the performance of the speech recognition module, which achieves reasonable results on clean read speech but degrades as this becomes more noisy and/or spontaneous.

A comparison using different speech parameters in the automatic emotion recognition using feature subset selection based on evolutionary algorithms

Study of emotions in human-computer interaction is a growing research area. Focusing on automatic emotion recognition, work is being performed in order to achieve good results particularly in speech and facial gesture recognition. This paper presents a study where, using a wide range of speech parameters, improvement in emotion recognition rates is analyzed. Using an emotional multimodal bilingual database for Spanish and Basque, emotion recognition rates in speech have significantly improved for both languages comparing with previous studies. In this particular case, as in previous studies, machine learning techniques based on evolutive algorithms (EDA) have proven to be the best emotion recognition rate optimizers.

Application of feature subset selection based on evolutionary algorithms for automatic emotion recognition in speech

The study of emotions in human-computer interaction is a growing research area. Focusing on automatic emotion recognition, work is being performed in order to achieve good results particularly in speech and facial gesture recognition. In this paper we present a study performed to analyze different machine learning techniques validity in automatic speech emotion recognition area. Using a bilingual affective database, different speech parameters have been calculated for each audio recording. Then, several machine learning techniques have been applied to evaluate their usefulness in speech emotion recognition, including techniques based on evolutive algorithms (EDA) to select speech feature subsets that optimize automatic emotion recognition success rate. Achieved experimental results show a representative increase in the success rate.

User Adapted Motor-Imaginary Brain-Computer Interface by means of EEG Channel Selection Based on Estimation of Distributed Algorithms

Brain-Computer Interfaces (BCIs) have become a research field with interesting applications, and it can be inferred from published papers that different persons activate different parts of the brain to perform the same action. This paper presents a personalized interface design method, for electroencephalogram- (EEG-) based BCIs, based on channel selection. We describe a novel two-step method in which firstly a computationally inexpensive greedy algorithm finds an adequate search range; and, then, an Estimation of Distribution Algorithm (EDA) is applied in the reduced range to obtain the optimal channel subset. The use of the EDA allows us to select the most interacting channels subset, removing the irrelevant and noisy ones, thus selecting the most discriminative subset of channels for each user improving accuracy. The method is tested on the IIIa dataset from the BCI competition III. Experimental results show that the resulting channel subset is consistent with motor-imaginary-related neurophysiological principles and, on the other hand, optimizes performance reducing the number of channels.

A real-time stress classification system based on arousal analysis of the nervous system by an F-state machine

BACKGROUND AND OBJECTIVE Detection and labelling of an increment in the human stress level is a contribution focused principally on improving the quality of life of people. This work is aimed to develop a biophysical real-time stress identification and classification system, analysing two noninvasive signals, the galvanic skin response and the heart rate variability. METHODS An experimental procedure was designed and configured in order to elicit a stressful situation that is similar to those found in real cases. A total of 166 subjects participated in this experimental stage. The set of registered signals of each subject was considered as one experiment. A preliminary qualitative analysis of the signals collected was made, based on previous counselling received from neurophysiologists and psychologists. This study revealed a relationship between changes in the temporal signals and the induced stress states in each subject. To identify and classify such states, a subsequent quantitative analysis was performed in order to determine specific numerical information related to the above mentioned relationship. This second analysis gives the particular details to design the finally proposed classification algorithm, based on a Finite State Machine. RESULTS The proposed system is able to classify the detected stress stages at three levels: low, medium, and high. Furthermore, the system identifies persistent stress situations or momentary alerts, depending on the subjects arousal. The system reaches an F1 score of 0.984 in the case of high level, an F1 score of 0.970 for medium level, and an F1 score of 0.943 for low level. CONCLUSION The resulting system is able to detect and classify different stress stages only based on two non invasive signals. These signals can be collected in people during their monitoring and be processed in a real-time sense, as the system can be previously preconfigured. Therefore, it could easily be implemented in a wearable prototype that could be worn by end users without feeling to be monitored. Besides, due to its low computational, the computation of the signals slopes is easy to do and its deployment in real-time applications is feasible.

Feature Selection for Speech Emotion Recognition in Spanish and Basque: On the Use of Machine Learning to Improve Human-Computer Interaction

Study of emotions in human–computer interaction is a growing research area. This paper shows an attempt to select the most significant features for emotion recognition in spoken Basque and Spanish Languages using different methods for feature selection. RekEmozio database was used as the experimental data set. Several Machine Learning paradigms were used for the emotion classification task. Experiments were executed in three phases, using different sets of features as classification variables in each phase. Moreover, feature subset selection was applied at each phase in order to seek for the most relevant feature subset. The three phases approach was selected to check the validity of the proposed approach. Achieved results show that an instance-based learning algorithm using feature subset selection techniques based on evolutionary algorithms is the best Machine Learning paradigm in automatic emotion recognition, with all different feature sets, obtaining a mean of 80,05% emotion recognition rate in Basque and a 74,82% in Spanish. In order to check the goodness of the proposed process, a greedy searching approach (FSS-Forward) has been applied and a comparison between them is provided. Based on achieved results, a set of most relevant non-speaker dependent features is proposed for both languages and new perspectives are suggested.