PEMO: A New Validated Dataset for Punjabi Speech Emotion Detection

This research work presents a new valid dataset for Punjabi called the Emotional Speech Database (PEMO) which has been developed to assess ability recognize emotions in speech by both computers and humans. The PEMO includes samples from about 60 speakers with an age range between 20 45 years, four fundamental emotions, including anger, sad, happy neutral. In order create data, films are retrieved different multimedia websites such as YouTube. movies processed transformed into utterances software PRAAT. database contains 22,000 natural utterances. is equivalent 12 hours 35 min of information taken online web series. Three annotators categorize emotional content common label that labelled all becomes final utterance. have thorough knowledge Language. data used determine expression