Acoustic Methods in Identifying Symptoms of Emotional States

The study investigates the use of speech signal to recognise speakers’ emotional states. introduction includes definition and categorization emotions, including facial expressions, physiological signals. For purpose this work, a proprietary resource emotionally-marked recordings was created. collected come from media, live journalistic broadcasts, which show spontaneous reactions real-time stimuli. analysis, specific script written in Python. Its algorithm parameterization determination features correlated with content speech. After parametrization process, data clustering performed allows for grouping feature vectors speakers into greater collections imitate Using t-Student test dependent samples, some descriptors were distinguished, identified significant differences values between Some potential applications research proposed, as well other development directions future studies topic.