Towards Interactive Agents that Infer Emotions from Voice and Context Information

Abstract

Conversational agents are increasingly being used for training of social skills. One of their most important benefits is their ability to understand the user`s emotions, to be able to provide natural interaction with humans. However, to infer a conversation partner's emotional state, humans typically make use of contextual information as well. This work proposes an architecture to extract emotions from human voice in combination with the context imprint of a particular situation. With that information, a computer system can achieve a more human-like type of interaction. The architecture presents satisfactory results. The strategy of combining 2 algorithms, one to cover 'common cases' and another to cover 'borderline cases' significantly reduces the percentage of mistakes in classification. The addition of context information also increases the accuracy in emotion inferences.

A serious game to improve the verbal resilience against doorstep scams

“…The SVM model was built using 4-fold cross validation over a dataset with 681 sentences of four people instructed in how to act into both categories. Details about the algorithm and the SVM tuning are described in [48]. The final accuracy of the module is 86.56%.…”

A Serious Game for Training Verbal Resilience to Doorstep Scams

“…These conditions are used by the speech When the speech analysis module is turned on, a few more steps are taken before continuing with the scenario. 5 Right after choosing a response (other than the repeat option) all the other responses disappear and a countdown is shown, after which the recording automatically starts. The player is instructed by the game to read their response aloud.…”

“…The SVM model was built using 4-fold cross validation over a dataset with 681 sentences of four people instructed in how to act into both categories. Details about the algorithm and the SVM tuning are described in [5]. The final accuracy of the module is 86.56%.…”