ITRW on Speech and Emotion September 5-7, 2000 Newcastle, Northern Ireland, UK

Automated extraction of ToBI annotation data from the Reading/Leeds emotional speech corpus

Richard Stibbard

Speech Research Laboratory, University of Reading, UK

In the interest of achieving experimental control, most studies of the effects of emotional arousal on the voice have avoided field data, preferring instead to work with data simulated by actors, or, more rarely due to the ethical and practical difficulties involved, induced in the laboratory. These methods allow for control over unwanted variables such as recording conditions, verbal content, accent, age, etc., but entail a corresponding loss of emotional realism (Murray & Arnott, 1993: 1100). Interest is now growing in improving psychological validity by using more natural speech samples in emotion research, despite the corresponding loss in experimental control.

Full Paper

Bibliographic reference.  Stibbard, Richard (2000): "Automated extraction of ToBI annotation data from the Reading/Leeds emotional speech corpus", In SpeechEmotion-2000, 60-65.