A frequently used procedure to examine the relationship between categorical and dimensional descriptions of emotions is to ask subjects to place verbal expressions representing emotions in a continuous multidimensional emotional space. This work chooses a different approach. It aims at creating a system predicting the values of Activation and Valence (AV) directly from the sound of emotional speech utterances without the use of its semantic content or any other additional information. The system uses X-vectors to represent sound characteristics of the utterance and Support Vector Regressor for the estimation the AV values. The system is trained on a pool of three publicly available databases with dimensional annotation of emotions. The quality of regression is evaluated on the test sets of the same databases. Mapping of categorical emotions to the dimensional space is tested on another pool of eight categorically annotated databases. The aim of the work was to test whether in each unseen database the predicted values of Valence and Activation will place emotion-tagged utterances in the AV space in accordance with expectations based on Russell’s circumplex model of affective space. Due to the great variability of speech data, clusters of emotions create overlapping clouds. Their average location can be represented by centroids. A hypothesis on the position of these centroids is formulated and evaluated. The system’s ability to separate the emotions is evaluated by measuring the distance of the centroids. It can be concluded that the system works as expected and the positions of the clusters follow the hypothesized rules. Although the variance in individual measurements is still very high and the overlap of emotion clusters is large, it can be stated that the AV coordinates predicted by the system lead to an observable separation of the emotions in accordance with the hypothesis. Knowledge from training databases can therefore be used to predict AV coordinates of unseen data of various origins. This could be used to detect high levels of stress or depression. With the appearance of more dimensionally annotated training data, the systems predicting emotional dimensions from speech sound will become more robust and usable in practical applications in call-centers, avatars, robots, information-providing systems, security applications, and the like.

Spoken communication between social robotic devices, powered by generative AI tools such as ChatGPT, and the senior population offers great potential for researching social interaction and robot identity perceptions as well as exploring the potential opportunities and challenges when implementing this human-machine interactions in real life situations and health care. In this paper we explore people’s perceptions of the social robot Furhat when administering verbal tasks similar to those used in screening for Alzheimer’s disease. We describe the Slovak system mounted on the robot that includes a speech recognizer, the scenarios powered by generative large language model ChatGPT, and a speech synthesizer. We tested the functionality of the proposed approach with two groups of participants: attendees of a large science fair and a scientific conference. The observations from 87 collected questionnaires suggest good potential and applicability of such an approach for the given task and more positive attitudes of older subjects compared to younger ones.

This study describes the methodology used for designing a database of speech under real stress. Based on limits of existing stress databases, we used a communication task via a computer game to collect speech data. To validate the presence of stress, known psychophysiological indicators such as heart rate and electrodermal activity, as well as subjective self-assessment were used. This paper presents the data from first 5 speakers (3 men, 2 women) who participated in initial tests of the proposed design. In 4 out of 5 speakers increases in fundamental frequency and intensity of speech were registered. Similarly, in 4 out of 5 speakers heart rate was significantly increased during the task, when compared with reference measurement from before the task. These first results show that proposed design might be appropriate for building a speech under stress database. However, there are still considerations that need to be addressed.

We propose a text classification method for the purpose of creating a language model for automatic recognition of spontaneous spoken speech. Transcripts from our departmental speech database served as spontaneous spoken texts. Using supervised machine learning methods, we have created multiple classification models (including neural networks), that were able to distinguish them from written texts with high accuracy. We subsequently verified the accuracy of our trained models on a database of texts containing direct speech extracted from newspaper articles.