Emotion identification using specific sentences that are biased towards their corresponding emotions

Speakers usually use certain words more frequently in expressing their emotions since they have learned the connection between certain words and their corresponding emotions. This work focuses on speaker-dependent and text-dependent emotion identification in completely two separate and different speech databases. One database uses neutral sentences that are unbiased towards any emotion; however, the second database uses certain sentences that are biased towards their corresponding emotions. Each database consists of six emotions: neutral, angry, sad, happy, disgust, and fear. Our results, based on hidden Markov models (HMMs), show that the emotion identification performance of the second database is much better than that of the first one.