ISCA Archive Interspeech 2008
ISCA Archive Interspeech 2008

Emotion conversion using F0 segment selection

Zeynep Inanoglu, Steve Young

This paper describes F0 segment selection, a novel syllable-based F0 conversion method, which provides a concatenative framework to search for F0 segments in a modest corpus of emotional speech (.15 minutes of data). The method is compared with our earlier work on F0 generation using context-sensitive syllable HMMs. Both methods are complemented with a duration conversion module as well as GMM-based spectral conversion to form a unified emotion conversion framework in English. The system was evaluated using three target styles: surprise, anger and sadness. The results of an extensive perceptual test show that segment selection significantly outperforms the HMM-based method in terms of both emotion recognition rates and intonation quality ratings for surprise and anger. For conveying sadness both methods were effective.

doi: 10.21437/Interspeech.2008-550

Cite as: Inanoglu, Z., Young, S. (2008) Emotion conversion using F0 segment selection. Proc. Interspeech 2008, 2122-2125, doi: 10.21437/Interspeech.2008-550

  author={Zeynep Inanoglu and Steve Young},
  title={{Emotion conversion using F0 segment selection}},
  booktitle={Proc. Interspeech 2008},