ISCA Archive Interspeech 2005
ISCA Archive Interspeech 2005

Automatic emotion recognition using prosodic parameters

Iker Luengo, Eva Navas, Inmaculada Hernáez, Jon Sánchez

This paper presents the experiments made to automatically identify emotion in an emotional speech database for Basque. Three different classifiers have been built: one using spectral features and GMM, other with prosodic features and SVM and the last one with prosodic features and GMM. 86 prosodic features were calculated and then an algorithm to select the most relevant ones was applied. The first classifier gives the best result with a 98.4% accuracy when using 512 mixtures, but the classifier built with the best 6 prosodic features achieves an accuracy of 92.3% in spite of its simplicity, showing that prosodic information is very useful to identify emotions.

doi: 10.21437/Interspeech.2005-324

Cite as: Luengo, I., Navas, E., Hernáez, I., Sánchez, J. (2005) Automatic emotion recognition using prosodic parameters. Proc. Interspeech 2005, 493-496, doi: 10.21437/Interspeech.2005-324

  author={Iker Luengo and Eva Navas and Inmaculada Hernáez and Jon Sánchez},
  title={{Automatic emotion recognition using prosodic parameters}},
  booktitle={Proc. Interspeech 2005},