Abstract

Over this decade the speech recognition plays an important role for speechmaker identification and identification of the various characteristics of a person involved in a particular section of the voice. The information obtained from those systems are used for interaction between the user and the machine. The emotion detection through face recognition needs to person’s face captured in the sensor for the detection of the emotion who are involved in the session. But voice recognition can be done without the knowledge of the that person. The application is used in the client service, education and health care. The high accuracy system of this type can be used in the intelligence services to know the truthfulness in investigation. For this task various parameters of the audio is obtained as feature called MFCC. The GMM modelling is successfully modelled by the feature MFCC. In addition to that UBM it gives better result in emotion of the speechmaker.