Quantification of Segmentation and F0 Errors and Their Effect on Emotion Recognition

Authors:
Stefan Steidl;Anton Batliner;Elmar Nöth;Joachim Hornegger
Affiliations:
Lehrstuhl für Mustererkennung, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany D-91058;Lehrstuhl für Mustererkennung, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany D-91058;Lehrstuhl für Mustererkennung, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany D-91058;Lehrstuhl für Mustererkennung, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany D-91058
Venue:
TSD '08 Proceedings of the 11th international conference on Text, Speech and Dialogue
Year:
2008

Citing 2
Cited 1

The SNNS Neural Network Simulator

Mustererkennung 1991, 13. DAGM-Symposium
How to find trouble in communication

Speech Communication - Special issue on speech and emotion

Recognising realistic emotions and affect in speech: State of the art and lessons learnt from the first challenge

Speech Communication

Quantified Score

Hi-index	0.00

Visualization

Abstract

Prosodic features modelling pitch, energy, and duration play a major role in speech emotion recognition. Our word level features, especially duration and pitch features, rely on correct word segmentation and F0 extraction. For the FAU Aibo Emotion Corpus, the automatic segmentation of a forced alignment of the spoken word sequence and the automatically extracted F0 values have been manually corrected. Frequencies of different types of segmentation and F0errors are given and their influence on emotion recognition using different groups of prosodic features is evaluated. The classification results show that the impact of these errors on emotion recognition is small.